Browse Electrical Engineering and Computer Science 2024 Research Projects
The LURA and NSERC USRA Summer 2024 Research Program competition is now closed. Applicants will be notified of results by April 1, 2024.
Professor: Ping Wang
Contact Info: pingw@yorku.ca
Lab Website: https://www.eecs.yorku.ca/~pingw/
Position Type: NSERC Undergraduate Student Research Award (USRA)
Open Positions: 1
Project Description: In the traditional machine learning (ML) approach, data collected by local nodes is uploaded to the data center and processed centrally, where formidable computational resources can be exploited. However, this approach is no longer suitable for edge computing networks. Firstly, users’ data is more privacy-sensitive than ever. Secondly, transferring data generated by many nodes for processing burdens the network and becomes the bottleneck of overall performance. Thirdly, the centric fashion involves a long propagation delay and incurs unacceptable latency, which is unbearable for many applications with instantaneous decision-making. Therefore, a natural question arises with the abovementioned concerns: How do we train a ML model from decentralized data at a resource-constrained edge node? Federated Learning (FL) is a technique that fulfills this purpose. FL is a ML setting where many nodes collaboratively train a model under the orchestration of a central server (e.g. service provider) while keeping training data decentralized. However, FL also faces challenges. One major challenge is system heterogeneity. FL involves the heterogeneous participants whose local dataset, computational ability, channel condition, power level and willingness to participate may vary. Given the system heterogeneity, an optimal strategy for resource allocation needs to be developed to maximize the efficiency of FL. Another challenge is communication costs during global model training. Compared with traditional distributed machine learning, where several computational centers are involved, the FL framework is usually related to many edge nodes, incurring an extremely high communication cost. Therefore, the effort to reduce communication costs (e.g., model compression) is needed to improve efficiency. This project will focus on developing cutting-edge FL techniques to address these challenges.
Duties and Responsibilities: The student will gain sufficient knowledge of federated learning and hands-on experience implementing federated learning algorithms. The student will do some research in the relevant field, with mentoring from senior graduate students.
Desired Technical Skills: Good at coding and have basic knowledge of machine learning.
Desired Course(s): Students from computer science are preferred.
Other Desired Qualifications: Good GPA and self-motivated.
Contact Info: pingw@yorku.ca
Lab Website: https://www.eecs.yorku.ca/~pingw/
Position Type: NSERC Undergraduate Student Research Award (USRA)
Open Positions: 1
Project Description: In the traditional machine learning (ML) approach, data collected by local nodes is uploaded to the data center and processed centrally, where formidable computational resources can be exploited. However, this approach is no longer suitable for edge computing networks. Firstly, users’ data is more privacy-sensitive than ever. Secondly, transferring data generated by many nodes for processing burdens the network and becomes the bottleneck of overall performance. Thirdly, the centric fashion involves a long propagation delay and incurs unacceptable latency, which is unbearable for many applications with instantaneous decision-making. Therefore, a natural question arises with the abovementioned concerns: How do we train a ML model from decentralized data at a resource-constrained edge node? Federated Learning (FL) is a technique that fulfills this purpose. FL is a ML setting where many nodes collaboratively train a model under the orchestration of a central server (e.g. service provider) while keeping training data decentralized. However, FL also faces challenges. One major challenge is system heterogeneity. FL involves the heterogeneous participants whose local dataset, computational ability, channel condition, power level and willingness to participate may vary. Given the system heterogeneity, an optimal strategy for resource allocation needs to be developed to maximize the efficiency of FL. Another challenge is communication costs during global model training. Compared with traditional distributed machine learning, where several computational centers are involved, the FL framework is usually related to many edge nodes, incurring an extremely high communication cost. Therefore, the effort to reduce communication costs (e.g., model compression) is needed to improve efficiency. This project will focus on developing cutting-edge FL techniques to address these challenges.
Duties and Responsibilities: The student will gain sufficient knowledge of federated learning and hands-on experience implementing federated learning algorithms. The student will do some research in the relevant field, with mentoring from senior graduate students.
Desired Technical Skills: Good at coding and have basic knowledge of machine learning.
Desired Course(s): Students from computer science are preferred.
Other Desired Qualifications: Good GPA and self-motivated.
Professor: Jerek Szlichta
Contact Info: szlichta@yorku.ca
Lab Website: https://www.yorku.ca/lassonde/lab/data-science
Position Type:
Open Positions: 2
Project Description: Modern data systems, such as IBM Db2, have dozens of system configuration parameters, commonly referred to as knobs. These parameters wield significant influence over the performance of business queries. Knobs are responsible for configuring various aspects, including the allocation of working memory, such as the number of pages allocated to the buffer pool and sortheap, the degree of parallelism to be used, and even toggle specific features by setting an optimization level.
Manual configuration tuning by experts is a labor-intensive and time-consuming process. Consequently, we propose XTune, a reliable and eXplainable, query-informed tuning system. XTune harnesses Deep Reinforcement Learning (DRL) techniques based on an actor-critic neural networks, specifically Proximal Policy Optimization (PPO), to tune system configurations. Notably, the PPO policy is considered state-of-the-art by OpenAI, owing to its stability, sample efficiency, and robustness in addressing various reinforcement learning challenges. It computes updates at each step to minimize the loss function while ensuring minimal deviation from the previous policy. The optimization process includes strategies like introducing back pressure to manage resource utilization in cloud computing for sustainability purposes. It begins with the translation of high-dimensional Query Execution Plans (QEPs) into a lower-dimensional space using embeddings derived from Bidirectional Encoder Representations from Transformers (BERT) and Graph Neural Networks (GNN), which then serve as inputs for the DRL models.
In the context of large-scale machine learning models, their inherent complexity often renders them as “black boxes,” posing challenges for experts to decipher their prediction processes. The lack of interpretability within predictive models undermines the confidence experts place in these models, particularly in scenarios involving critical decisions, such data systems tuning. To tackle this issue and cultivate enhanced interpretability within data systems, our research introduces methods to generate saliency and counterfactual explanations, effectively transforming these black boxes into “glass boxes” that offer individuals insights into their internal mechanisms. Our saliency explanation method for tuning system configurations approximates the importance of model features, such as query subplans. On the other hand, our counterfactual explanations reveal what should have been different in queries and Query Execution Plans (QEPs) in terms of perturbations to observe a diverse or desired outcome. To further enhance our approach, we implement an instance-based counterfactual strategy. This strategy outputs similar QEPs from the workload, rather than using arbitrary perturbations, resulting in a diverse tuning outcome.
We evaluate our methods over synthetic and real query workloads, quantifying their effectiveness and performance benefits, particularly in the context of data lake-driven workloads. The development of XTune advances the reliability of data systems, while also aligning with the principles of sustainability, resulting in responsible technology usage. Ultimately, the impact of XTune resonates across industries, illustrating how responsible AI can drive positive change.
Duties and Responsibilities: Students’ duties and responsibilities will include: reviewing related work in automatic knobs tuning for data systems, designing large-scale machine learning-driven approaches to the tuning of configuration parameters, implementing the solution with the deep reinforcement learning model, conducting comprehensive experimental evaluation over synthetic and real-world query workloads, and writing a research paper to be submitted to one of the top-tier conferences in data science, such as VLDB, ACM SIGMOD, IEEE ICDE and EDBT.
Desired Technical Skills: The students should possess algorithmic design and development knowledge, as well as demonstrated strong programming skills.
Desired Course(s): It is recommended to have completed some of the data science courses such as LE/EECS 4415 3.00 – Big Data Systems, LE/EECS 4411 3.00 – Database Management Systems, LE/EECS 4412 3.00 – Data Mining, LE/EECS 4404 3.00 – Introduction to Machine Learning and Pattern Recognition, LE/EECS 3421 3.00 – Introduction to Database Systems, etc.
Other Desired Qualifications: Other qualifications include good communication skills.
Contact Info: szlichta@yorku.ca
Lab Website: https://www.yorku.ca/lassonde/lab/data-science
Position Type:
Open Positions: 2
Project Description: Modern data systems, such as IBM Db2, have dozens of system configuration parameters, commonly referred to as knobs. These parameters wield significant influence over the performance of business queries. Knobs are responsible for configuring various aspects, including the allocation of working memory, such as the number of pages allocated to the buffer pool and sortheap, the degree of parallelism to be used, and even toggle specific features by setting an optimization level.
Manual configuration tuning by experts is a labor-intensive and time-consuming process. Consequently, we propose XTune, a reliable and eXplainable, query-informed tuning system. XTune harnesses Deep Reinforcement Learning (DRL) techniques based on an actor-critic neural networks, specifically Proximal Policy Optimization (PPO), to tune system configurations. Notably, the PPO policy is considered state-of-the-art by OpenAI, owing to its stability, sample efficiency, and robustness in addressing various reinforcement learning challenges. It computes updates at each step to minimize the loss function while ensuring minimal deviation from the previous policy. The optimization process includes strategies like introducing back pressure to manage resource utilization in cloud computing for sustainability purposes. It begins with the translation of high-dimensional Query Execution Plans (QEPs) into a lower-dimensional space using embeddings derived from Bidirectional Encoder Representations from Transformers (BERT) and Graph Neural Networks (GNN), which then serve as inputs for the DRL models.
In the context of large-scale machine learning models, their inherent complexity often renders them as “black boxes,” posing challenges for experts to decipher their prediction processes. The lack of interpretability within predictive models undermines the confidence experts place in these models, particularly in scenarios involving critical decisions, such data systems tuning. To tackle this issue and cultivate enhanced interpretability within data systems, our research introduces methods to generate saliency and counterfactual explanations, effectively transforming these black boxes into “glass boxes” that offer individuals insights into their internal mechanisms. Our saliency explanation method for tuning system configurations approximates the importance of model features, such as query subplans. On the other hand, our counterfactual explanations reveal what should have been different in queries and Query Execution Plans (QEPs) in terms of perturbations to observe a diverse or desired outcome. To further enhance our approach, we implement an instance-based counterfactual strategy. This strategy outputs similar QEPs from the workload, rather than using arbitrary perturbations, resulting in a diverse tuning outcome.
We evaluate our methods over synthetic and real query workloads, quantifying their effectiveness and performance benefits, particularly in the context of data lake-driven workloads. The development of XTune advances the reliability of data systems, while also aligning with the principles of sustainability, resulting in responsible technology usage. Ultimately, the impact of XTune resonates across industries, illustrating how responsible AI can drive positive change.
Duties and Responsibilities: Students’ duties and responsibilities will include: reviewing related work in automatic knobs tuning for data systems, designing large-scale machine learning-driven approaches to the tuning of configuration parameters, implementing the solution with the deep reinforcement learning model, conducting comprehensive experimental evaluation over synthetic and real-world query workloads, and writing a research paper to be submitted to one of the top-tier conferences in data science, such as VLDB, ACM SIGMOD, IEEE ICDE and EDBT.
Desired Technical Skills: The students should possess algorithmic design and development knowledge, as well as demonstrated strong programming skills.
Desired Course(s): It is recommended to have completed some of the data science courses such as LE/EECS 4415 3.00 – Big Data Systems, LE/EECS 4411 3.00 – Database Management Systems, LE/EECS 4412 3.00 – Data Mining, LE/EECS 4404 3.00 – Introduction to Machine Learning and Pattern Recognition, LE/EECS 3421 3.00 – Introduction to Database Systems, etc.
Other Desired Qualifications: Other qualifications include good communication skills.
Professor: Alvine Boaye Belle
Contact Info: alvine.belle@lassonde.yorku.ca
Lab Website: https://lassonde.yorku.ca/users/alvinebelle
Position Type:
Open Positions: 2
Project Description: The number of citations of scientific articles has a huge impact on recommendations for funding allocations, recruitment decisions, and rewards, just to name a few. However, some researchers belonging to some socio-cultural groups (e.g., women) are usually less cited than other researchers coming from dominating groups. This may be due to the presence of some unfairness citation patterns in some scientific articles. These citation patterns are tangible examples of biases against researchers from some socio-cultural groups and may inevitably cause unfairness and inaccuracy in the assessment of articles impact. These citation patterns may therefore translate to significant disparities in promotion, retention, grant funding, awards, collaborative opportunities, and publications. The project will first start by analyzing the existing scientific literature to find out the various unfairness citations patterns that may be present in some scientific articles. Then, the project will focus on the exploration of existing mitigation solutions and their limitations. The project will then aim at developing an online tool called CiteFair that will be able to: 1. Automatically analyze scientific articles to detect the potential presence of unfairness citation patterns. 2. Rely on existing bibliometric tools to provide some suggestions to articles authors to mitigate these citations patterns and increase the fairness citation score of their articles.
The project will also consist in validating the accuracy of the CiteFair tool by making experiments on a sample of the scientific articles published within the last decade in a wide range of venues. Experiments will also focus on evaluating the usability and performance of the CiteFair tool.
Duties and Responsibilities: 1. Read scientific papers to become more knowledgeable of unfairness citation patterns. 2. Design an online tool called CiteFair, and able to automatically analyze scientific articles to detect the potential presence of unfairness citation patterns. 3. Use web technologies to implement the designed tool. 4. Conduct experiments to assess the accuracy of the CiteFair tool.
Desired Technical Skills: Solid experience with JavaScript, HTML, and CSS, and good experience with web-development frameworks (e.g., React JS, Spring Boot).
Desired Course(s): Priority will be given to students who have already taken LE/EECS 4413 3.00 – Building e-commerce systems or who have already developed online tools.
Other Desired Qualifications: Good oral and written skills in English.
Contact Info: alvine.belle@lassonde.yorku.ca
Lab Website: https://lassonde.yorku.ca/users/alvinebelle
Position Type:
Open Positions: 2
Project Description: The number of citations of scientific articles has a huge impact on recommendations for funding allocations, recruitment decisions, and rewards, just to name a few. However, some researchers belonging to some socio-cultural groups (e.g., women) are usually less cited than other researchers coming from dominating groups. This may be due to the presence of some unfairness citation patterns in some scientific articles. These citation patterns are tangible examples of biases against researchers from some socio-cultural groups and may inevitably cause unfairness and inaccuracy in the assessment of articles impact. These citation patterns may therefore translate to significant disparities in promotion, retention, grant funding, awards, collaborative opportunities, and publications. The project will first start by analyzing the existing scientific literature to find out the various unfairness citations patterns that may be present in some scientific articles. Then, the project will focus on the exploration of existing mitigation solutions and their limitations. The project will then aim at developing an online tool called CiteFair that will be able to: 1. Automatically analyze scientific articles to detect the potential presence of unfairness citation patterns. 2. Rely on existing bibliometric tools to provide some suggestions to articles authors to mitigate these citations patterns and increase the fairness citation score of their articles.
The project will also consist in validating the accuracy of the CiteFair tool by making experiments on a sample of the scientific articles published within the last decade in a wide range of venues. Experiments will also focus on evaluating the usability and performance of the CiteFair tool.
Duties and Responsibilities: 1. Read scientific papers to become more knowledgeable of unfairness citation patterns. 2. Design an online tool called CiteFair, and able to automatically analyze scientific articles to detect the potential presence of unfairness citation patterns. 3. Use web technologies to implement the designed tool. 4. Conduct experiments to assess the accuracy of the CiteFair tool.
Desired Technical Skills: Solid experience with JavaScript, HTML, and CSS, and good experience with web-development frameworks (e.g., React JS, Spring Boot).
Desired Course(s): Priority will be given to students who have already taken LE/EECS 4413 3.00 – Building e-commerce systems or who have already developed online tools.
Other Desired Qualifications: Good oral and written skills in English.
Professor: Zhen Ming (Jack) Jiang
Contact Info: zmjiang@yorku.ca
Lab Website: http://www.cse.yorku.ca/~zmjiang/
Position Type: Lassonde Undergraduate Research Award (LURA); NSERC Undergraduate Student Research Award (USRA)
Open Positions: 2
Project Description: Software engineering data (e.g., source code repositories and bug databases) contain a wealth of information about a project’s status and history. With the recent advances of large language models (e.g., GPT and BERT) as well as their applications (e.g., ChatGPT or GitHub Copilot), many software engineering tasks can be automated or optimized. In this project, the student(s) will explore and investigate various software engineering applications which can benefit from the use of LLMs.
Duties and Responsibilities: Conducting research, implementing research prototype, and evaluating their tools.
Desired Technical Skills: Proficient programming in Python and Java, some knowledge in AI and Machine Learning, and have tried some large language models and/or their applications (e.g., ChatGPT or Copilot).
Desired Course(s): Third year or above in Computer Science/Computer Engineering/Software Engineering, and at least B+ in LE/EECS 3311 3.00 – Software Design.
Other Desired Qualifications: Ability to learn things quickly, and good communication skills.
Contact Info: zmjiang@yorku.ca
Lab Website: http://www.cse.yorku.ca/~zmjiang/
Position Type: Lassonde Undergraduate Research Award (LURA); NSERC Undergraduate Student Research Award (USRA)
Open Positions: 2
Project Description: Software engineering data (e.g., source code repositories and bug databases) contain a wealth of information about a project’s status and history. With the recent advances of large language models (e.g., GPT and BERT) as well as their applications (e.g., ChatGPT or GitHub Copilot), many software engineering tasks can be automated or optimized. In this project, the student(s) will explore and investigate various software engineering applications which can benefit from the use of LLMs.
Duties and Responsibilities: Conducting research, implementing research prototype, and evaluating their tools.
Desired Technical Skills: Proficient programming in Python and Java, some knowledge in AI and Machine Learning, and have tried some large language models and/or their applications (e.g., ChatGPT or Copilot).
Desired Course(s): Third year or above in Computer Science/Computer Engineering/Software Engineering, and at least B+ in LE/EECS 3311 3.00 – Software Design.
Other Desired Qualifications: Ability to learn things quickly, and good communication skills.
Professor: Hamzeh Khazaei
Contact Info: hkh@yorku.ca
Lab Website: https://pacslab.ca
Position Type: NSERC Undergraduate Student Research Award (USRA)
Open Positions: 2
Project Description: Large-scale distributed systems are becoming more challenging to design, implement, operate, and maintain. In particular, manual run-time management of distributed cloud systems is no longer an option. As a result, we need to leverage AI/ML techniques to help operate these large-scale systems aiming for the highest availability and reliability. This will be even more important when we use these systems in critical sectors such as health. In the PACS lab, we are working on designing, implementing, and prototyping AI/ML techniques to make distributed systems smarter so that they can take care of themselves to a high degree. The two prospective undergraduate students will help develop and prototype such intelligent distributed cloud systems during the summer.
Duties and Responsibilities: Brief background reading will be required, followed by work on the development of new AI techniques to provide smarter cloud-distributed systems.
Desired Technical Skills: Familiarity with cloud technologies (e.g. Docker, Microservices and FaaS), basic knowledge in ML and AI and programming in Python or Go languages.
Desired Course(s): Computer science or software engineering students who have received good grades in LE/EECS 2011 3.00 – Fundamentals of Data Structures and/or LE/EECS 3221 3.00 – Operating System Fundamentals.
Other Desired Qualifications: Interested in large-scale distributed systems, cloud computing and ML/AI.
Contact Info: hkh@yorku.ca
Lab Website: https://pacslab.ca
Position Type: NSERC Undergraduate Student Research Award (USRA)
Open Positions: 2
Project Description: Large-scale distributed systems are becoming more challenging to design, implement, operate, and maintain. In particular, manual run-time management of distributed cloud systems is no longer an option. As a result, we need to leverage AI/ML techniques to help operate these large-scale systems aiming for the highest availability and reliability. This will be even more important when we use these systems in critical sectors such as health. In the PACS lab, we are working on designing, implementing, and prototyping AI/ML techniques to make distributed systems smarter so that they can take care of themselves to a high degree. The two prospective undergraduate students will help develop and prototype such intelligent distributed cloud systems during the summer.
Duties and Responsibilities: Brief background reading will be required, followed by work on the development of new AI techniques to provide smarter cloud-distributed systems.
Desired Technical Skills: Familiarity with cloud technologies (e.g. Docker, Microservices and FaaS), basic knowledge in ML and AI and programming in Python or Go languages.
Desired Course(s): Computer science or software engineering students who have received good grades in LE/EECS 2011 3.00 – Fundamentals of Data Structures and/or LE/EECS 3221 3.00 – Operating System Fundamentals.
Other Desired Qualifications: Interested in large-scale distributed systems, cloud computing and ML/AI.
Professor: Robert Allison
Contact Info: allison@eecs.yorku.ca
Lab Website: https://percept.eecs.yorku.ca/
Position Type: Lassonde Undergraduate Research Award (LURA); NSERC Undergraduate Student Research Award (USRA)
Open Positions: 2
Project Description: Students will help design, develop and conduct experiments related to human-computer interaction in virtual environments and digital media. In our lab we have a wide range of apparatuses to study human perception in computer-mediated worlds including a new and unique fully immersive virtual environment display. The student will develop interactive 3D virtual worlds and conduct experiments to study self-motion perception, visual perception, and human computer interaction in these virtual worlds. In particular, working with a senior graduate student or postdoctoral fellow, the successful applicant will model 3D environments, render them in a virtual reality or other digital media display, develop/implement interaction methods to control and interact with the simulation, and/or develop and run experimental scenarios to investigate these issues with human participants.
Duties and Responsibilities: Depending on skills and preparation the student would be responsible for: literature reviews and research, design of virtual environments, computer programming, testing, recruiting participants, conducting user studies, modelling and data analysis, preparation of reports, graphics and presentations.
Desired Technical Skills: Good programming skills, previous work with computer graphics or virtual reality would be helpful, as would basic mechanical skills. Students with background in psychology and interest in experimental psychology are also welcome to apply. Artistic background or skill would be an asset but is not required.
Desired Course(s): Digital media, electrical engineering, computer engineering, computer science, psychology, or vision science students.
Other Desired Qualifications: Students should be self-directed and work well in a team environment.
Contact Info: allison@eecs.yorku.ca
Lab Website: https://percept.eecs.yorku.ca/
Position Type: Lassonde Undergraduate Research Award (LURA); NSERC Undergraduate Student Research Award (USRA)
Open Positions: 2
Project Description: Students will help design, develop and conduct experiments related to human-computer interaction in virtual environments and digital media. In our lab we have a wide range of apparatuses to study human perception in computer-mediated worlds including a new and unique fully immersive virtual environment display. The student will develop interactive 3D virtual worlds and conduct experiments to study self-motion perception, visual perception, and human computer interaction in these virtual worlds. In particular, working with a senior graduate student or postdoctoral fellow, the successful applicant will model 3D environments, render them in a virtual reality or other digital media display, develop/implement interaction methods to control and interact with the simulation, and/or develop and run experimental scenarios to investigate these issues with human participants.
Duties and Responsibilities: Depending on skills and preparation the student would be responsible for: literature reviews and research, design of virtual environments, computer programming, testing, recruiting participants, conducting user studies, modelling and data analysis, preparation of reports, graphics and presentations.
Desired Technical Skills: Good programming skills, previous work with computer graphics or virtual reality would be helpful, as would basic mechanical skills. Students with background in psychology and interest in experimental psychology are also welcome to apply. Artistic background or skill would be an asset but is not required.
Desired Course(s): Digital media, electrical engineering, computer engineering, computer science, psychology, or vision science students.
Other Desired Qualifications: Students should be self-directed and work well in a team environment.
Professor: Pooja Vashisth
Contact Info: vashistp@yorku.ca
Lab Website: https://lassonde.yorku.ca/users/pvashisth
Position Type: Lassonde Undergraduate Research Award (LURA)
Open Positions: 1
Project Description: Problems and Needs: Moodle – York University Learning Management System (LMS) – provides instructors with rich data sets for students’ activities and performance. However, while the data comes in bulk, some important information (e.g. course activities) is automatically deleted by the system and replaced with new data each week. This issue prevents instructors from using such data effectively to enhance their teaching. Moreover, the bulk data may prevent professors from taking away useful insights to improve course quality.
Ideas: This project aims to address the main problems below with the following solutions: Retrieve full data set using scripts/integrations/apps that automatically pull course data from Moodle. Provide instructors with useful data visualizations from Moodle’s enormous data to support their teaching. Provide a quick summary and insights from those data sets. Hence based on the insights generated by the application, instructors can be supported with an early intervention system.
Implementation:
Pull data – quiz statistics, proposal, weekly course activity, new analytics → report, course activity, course grades.
Data visualization: visualize general data, data of assessments (quizzes, assignments,…), summary for all assessments and summary for each assessment and for each question, mean, mode, median, average time overall and for each assessment, average number of attempts, grade distribution, time distribution (when and how long students spend on) → clusters, respondents distribution (based on best attempt and compared to class) for each question, correlation between the number of attempts and accuracy, highlight hard questions and corresponding materials/reading views and engagement, extract question labels/keywords, what are the types of questions (MCQ, pseudocode,…) that students are generally not performing well on, data of exam, mean, mode, median, grade distribution, data of engagement, time distribution (when and how long students spend on) per course material, most and least viewed/engaged materials, average views and engagement per material, is there any correlation between students’ engagement and performance (i.e. do students actually achieve higher scores if they engage with course materials more frequently), visualize individual data, current avg. mark, all marks of that student up to that moment, performance as a graph (quizzes, assignments, test 1, test 2, final), attempts and accuracy per quiz, time spent on each assessment, time a student started an assessment and the last submission timestamp
Proposal 3, questions/topics that each student is doing well/struggling with engagement: what content does this student engage with? how often? when? how long? times opened a document? correlation between engagement and performance? feedback and insights, individual feedback – assess whether a student is at risk or not, if a student does well on assignments, how do they perform on midterms and finals? how about the opposite case? [tentative] predict examination results? interventions, when and how instructors should deploy their interventions to students, based on the results provided? [tentative] option to email/inform those low-performing students?
Deliverables: website to show statistical visualizations (high priority), sign in with Moodle, all students’ data graphs individuals’ data graphs feedback by words | insights for each student, automation on data collection from Moodle (medium priority) instructors will be able to share those results with students (optional – low priority).
Duties and Responsibilities: Deliverables: website to show statistical visualizations (high priority), sign in with Moodle, all students’ data graphs individuals’ data graphs feedback by words | insights for each student, automation on data collection from Moodle (medium priority) instructors will be able to share those results with students (optional – low priority).
Desired Technical Skills: Research and statistics skills; web development (applying OOP design, design pattern, and SOLID principles in developing backend API (using Python Flask)); data analysis: using Python framework and library (pandas, NumPy, seaborn, matplotlib, statsmodel) in analyzing the data; full-stack development: front-end: building user interfaces with JavaScript technologies; back-end: implementing RESTful API and GraphQL API using JavaScript technologies; DBMS: database design and using NoSQL/SQL databases in web development.
Desired Course(s): The position is open for third and fourth year students in a computer science, statistics, data science, or engineering degree.
Other Desired Qualifications: Good at programming, statistics, research, and has a willingness to explore the unknown.
Contact Info: vashistp@yorku.ca
Lab Website: https://lassonde.yorku.ca/users/pvashisth
Position Type: Lassonde Undergraduate Research Award (LURA)
Open Positions: 1
Project Description: Problems and Needs: Moodle – York University Learning Management System (LMS) – provides instructors with rich data sets for students’ activities and performance. However, while the data comes in bulk, some important information (e.g. course activities) is automatically deleted by the system and replaced with new data each week. This issue prevents instructors from using such data effectively to enhance their teaching. Moreover, the bulk data may prevent professors from taking away useful insights to improve course quality.
Ideas: This project aims to address the main problems below with the following solutions: Retrieve full data set using scripts/integrations/apps that automatically pull course data from Moodle. Provide instructors with useful data visualizations from Moodle’s enormous data to support their teaching. Provide a quick summary and insights from those data sets. Hence based on the insights generated by the application, instructors can be supported with an early intervention system.
Implementation:
Pull data – quiz statistics, proposal, weekly course activity, new analytics → report, course activity, course grades.
Data visualization: visualize general data, data of assessments (quizzes, assignments,…), summary for all assessments and summary for each assessment and for each question, mean, mode, median, average time overall and for each assessment, average number of attempts, grade distribution, time distribution (when and how long students spend on) → clusters, respondents distribution (based on best attempt and compared to class) for each question, correlation between the number of attempts and accuracy, highlight hard questions and corresponding materials/reading views and engagement, extract question labels/keywords, what are the types of questions (MCQ, pseudocode,…) that students are generally not performing well on, data of exam, mean, mode, median, grade distribution, data of engagement, time distribution (when and how long students spend on) per course material, most and least viewed/engaged materials, average views and engagement per material, is there any correlation between students’ engagement and performance (i.e. do students actually achieve higher scores if they engage with course materials more frequently), visualize individual data, current avg. mark, all marks of that student up to that moment, performance as a graph (quizzes, assignments, test 1, test 2, final), attempts and accuracy per quiz, time spent on each assessment, time a student started an assessment and the last submission timestamp
Proposal 3, questions/topics that each student is doing well/struggling with engagement: what content does this student engage with? how often? when? how long? times opened a document? correlation between engagement and performance? feedback and insights, individual feedback – assess whether a student is at risk or not, if a student does well on assignments, how do they perform on midterms and finals? how about the opposite case? [tentative] predict examination results? interventions, when and how instructors should deploy their interventions to students, based on the results provided? [tentative] option to email/inform those low-performing students?
Deliverables: website to show statistical visualizations (high priority), sign in with Moodle, all students’ data graphs individuals’ data graphs feedback by words | insights for each student, automation on data collection from Moodle (medium priority) instructors will be able to share those results with students (optional – low priority).
Duties and Responsibilities: Deliverables: website to show statistical visualizations (high priority), sign in with Moodle, all students’ data graphs individuals’ data graphs feedback by words | insights for each student, automation on data collection from Moodle (medium priority) instructors will be able to share those results with students (optional – low priority).
Desired Technical Skills: Research and statistics skills; web development (applying OOP design, design pattern, and SOLID principles in developing backend API (using Python Flask)); data analysis: using Python framework and library (pandas, NumPy, seaborn, matplotlib, statsmodel) in analyzing the data; full-stack development: front-end: building user interfaces with JavaScript technologies; back-end: implementing RESTful API and GraphQL API using JavaScript technologies; DBMS: database design and using NoSQL/SQL databases in web development.
Desired Course(s): The position is open for third and fourth year students in a computer science, statistics, data science, or engineering degree.
Other Desired Qualifications: Good at programming, statistics, research, and has a willingness to explore the unknown.
Professor: Simone Pisana
Contact Info: pisana@yorku.ca
Lab Website: https://pisana.lab.yorku.ca/
Position Type: Lassonde Undergraduate Research Award (LURA); NSERC Undergraduate Student Research Award (USRA)
Open Positions: 1
Project Description: We are starting a new project that aims at using graphene membranes to separate hydrogen isotopes in a solution. Graphene is a one-atom thick layer of carbon atoms in a honeycomb crystal. Despite its atomic thickness, it is impermeable to all but the smallest atoms, in fact only hydrogen ions (protons) have an appreciable transport across it. The transport goes down if the hydrogen is “heavier”, a deuterium nucleus. By sandwiching graphene and fuel cell membranes and applying a voltage, one can produce an electrochemical pump that transports protons. it was shown that such composite membranes can behave as a filter to enrich and separate heavy water (water composed of deuterium atoms instead of hydrogen) from normal water. We would like to fabricate these devices and test them for potential applications to the nuclear industry.
Duties and Responsibilities: The successful candidate will be responsible of reviewing the published methods to fabricate proton electrochemical pumps, fabricate the structures and characterize them in the laboratory.
Desired Technical Skills: Independent thinker and investigator, can think outside the box and present solutions to a problem. Comfortable with applying skills practically, having done so in the past (i.e. D.I.Y. and passion projects). Knowledge of crystal and atomic structure, chemistry of materials.
Desired Course(s): No particular degree program required, but the candidate will have passed with good grades fundamental courses in physics and chemistry.
Other Desired Qualifications: N/A.
Contact Info: pisana@yorku.ca
Lab Website: https://pisana.lab.yorku.ca/
Position Type: Lassonde Undergraduate Research Award (LURA); NSERC Undergraduate Student Research Award (USRA)
Open Positions: 1
Project Description: We are starting a new project that aims at using graphene membranes to separate hydrogen isotopes in a solution. Graphene is a one-atom thick layer of carbon atoms in a honeycomb crystal. Despite its atomic thickness, it is impermeable to all but the smallest atoms, in fact only hydrogen ions (protons) have an appreciable transport across it. The transport goes down if the hydrogen is “heavier”, a deuterium nucleus. By sandwiching graphene and fuel cell membranes and applying a voltage, one can produce an electrochemical pump that transports protons. it was shown that such composite membranes can behave as a filter to enrich and separate heavy water (water composed of deuterium atoms instead of hydrogen) from normal water. We would like to fabricate these devices and test them for potential applications to the nuclear industry.
Duties and Responsibilities: The successful candidate will be responsible of reviewing the published methods to fabricate proton electrochemical pumps, fabricate the structures and characterize them in the laboratory.
Desired Technical Skills: Independent thinker and investigator, can think outside the box and present solutions to a problem. Comfortable with applying skills practically, having done so in the past (i.e. D.I.Y. and passion projects). Knowledge of crystal and atomic structure, chemistry of materials.
Desired Course(s): No particular degree program required, but the candidate will have passed with good grades fundamental courses in physics and chemistry.
Other Desired Qualifications: N/A.
Professor: Peter Lian
Contact Info: peterlian@eecs.yorku.ca
Lab Website: https://lassonde.yorku.ca/users/peterlian
Position Type: Lassonde Undergraduate Research Award (LURA); NSERC Undergraduate Student Research Award (USRA)
Open Positions: 4
Project Description: COVID-19 that is caused by severe acute respiratory syndrome coronavirus 2 has affected our daily life for the past three years with large number of confirmed cases and deaths globally. Real-time monitoring of COVID-19 infection could be an effective way to help containing the COVID-19 pandemic and managing post COVID-19 recovery. This project aims to develop wireless flexible biomedical sensors that are able to track physiological and behavioural changes associated with COVID-19 viral infection. A flexible sensor is a thin and light-weight wearable device that conforms to body shape, which is very similar to bandage tape. It can be used to record human vital signs such as respiratory rate, blood oxygen level, electrocardiogram results, blood pressure, and skin temperature in real-time, i.e. 24 hours a day, 7 days a week. Such sensors are more acceptable by users and patients, which help improve health care compliance. It can also provide real-time information remotely to doctors when a user or a patient needs attention or help. The project consists of four parts: hardware development, firmware, a machine learning algorithm, and the user interface. Each part is carried out by one or more students. The hardware development includes the design of circuit board in flexible PCB, system tests, and integration. The firmware involves the coding for microcontrollers and Bluetooth wireless transceivers. The machine learning algorithm aims to implement algorithms to extract important features from recorded vital signs. The user interface is an app for smartphones that allows user to easily interact with the sensor.
The student will work at the Smart Sensor Lab from the EECS department. This project is part of the NSERC industrial stream CREATE-Microsystems Technologies & Application (MTA) training program. Please refer to https://www.yorku.ca/research/project/mta/ for more details of CREATE-MTA.
Duties and Responsibilities: This is a group project. The successful candidates will be responsible for one of the followings: Student 1: the design of PCB and building circuit board Student 2: firmware development Student 3: machine learning algorithm and neural network training, and Student 4: app development.
Desired Technical Skills: The student should have one of following qualifications: Basic knowledge of circuits and PCB design, knowledge in microcontroller programming, knowledge in Bluetooth is a plus, basic knowledge of digital signal processing, knowledge in machine learning is a plus, good knowledge in programming and app development, passionate about hardware development or programming.
Desired Course(s): Minimum requirement: LE/EECS 2200 3.00 – Electrical Circuits, LE/EECS 2210 3.00 – Electronic Circuits and Devices or equivalent for hardware part, LE/EECS 2030 3.00 – Advanced Object Oriented Programming, LE/EECS 2031 3.00 – Software Tools or equivalent for software part. Electrical engineering, computer engineering, and computer science students are all welcome.
Other Desired Qualifications: Experience with PCB design, Bluetooth, machine learning algorithm, and app development.
Contact Info: peterlian@eecs.yorku.ca
Lab Website: https://lassonde.yorku.ca/users/peterlian
Position Type: Lassonde Undergraduate Research Award (LURA); NSERC Undergraduate Student Research Award (USRA)
Open Positions: 4
Project Description: COVID-19 that is caused by severe acute respiratory syndrome coronavirus 2 has affected our daily life for the past three years with large number of confirmed cases and deaths globally. Real-time monitoring of COVID-19 infection could be an effective way to help containing the COVID-19 pandemic and managing post COVID-19 recovery. This project aims to develop wireless flexible biomedical sensors that are able to track physiological and behavioural changes associated with COVID-19 viral infection. A flexible sensor is a thin and light-weight wearable device that conforms to body shape, which is very similar to bandage tape. It can be used to record human vital signs such as respiratory rate, blood oxygen level, electrocardiogram results, blood pressure, and skin temperature in real-time, i.e. 24 hours a day, 7 days a week. Such sensors are more acceptable by users and patients, which help improve health care compliance. It can also provide real-time information remotely to doctors when a user or a patient needs attention or help. The project consists of four parts: hardware development, firmware, a machine learning algorithm, and the user interface. Each part is carried out by one or more students. The hardware development includes the design of circuit board in flexible PCB, system tests, and integration. The firmware involves the coding for microcontrollers and Bluetooth wireless transceivers. The machine learning algorithm aims to implement algorithms to extract important features from recorded vital signs. The user interface is an app for smartphones that allows user to easily interact with the sensor.
The student will work at the Smart Sensor Lab from the EECS department. This project is part of the NSERC industrial stream CREATE-Microsystems Technologies & Application (MTA) training program. Please refer to https://www.yorku.ca/research/project/mta/ for more details of CREATE-MTA.
Duties and Responsibilities: This is a group project. The successful candidates will be responsible for one of the followings: Student 1: the design of PCB and building circuit board Student 2: firmware development Student 3: machine learning algorithm and neural network training, and Student 4: app development.
Desired Technical Skills: The student should have one of following qualifications: Basic knowledge of circuits and PCB design, knowledge in microcontroller programming, knowledge in Bluetooth is a plus, basic knowledge of digital signal processing, knowledge in machine learning is a plus, good knowledge in programming and app development, passionate about hardware development or programming.
Desired Course(s): Minimum requirement: LE/EECS 2200 3.00 – Electrical Circuits, LE/EECS 2210 3.00 – Electronic Circuits and Devices or equivalent for hardware part, LE/EECS 2030 3.00 – Advanced Object Oriented Programming, LE/EECS 2031 3.00 – Software Tools or equivalent for software part. Electrical engineering, computer engineering, and computer science students are all welcome.
Other Desired Qualifications: Experience with PCB design, Bluetooth, machine learning algorithm, and app development.
Professor: Yan Shvartzshnaider
Contact Info: rhythm.lab@yorku.ca
Lab Website: https://www.yorku.ca/lassonde/privacy/
Position Type: NSERC Undergraduate Student Research Award (USRA)
Open Positions: 2
Project Description: Modern sociotechnical systems share and collect vast amounts of information. These systems violate users’ privacy by ignoring the context in which the information is shared and by implementing privacy models that fail to incorporate contextual information norms.
Using techniques in natural language processing, machine learning, network, and data analysis, this project is set to explore the privacy implications of mobile apps, online platforms, and other systems in different social contexts/settings.
To tackle this challenge, the project will operationalize a cutting-edge privacy theory and methodologies to conduct an analysis of existing technologies and design privacy-enhancing tools.
Duties and Responsibilities: Students will develop privacy-enhancing mechanisms that ensure that information flows in accordance with users’ expectations and established societal norms and design. Specific tasks include: comprehensive literature review of existing methodologies and tools, analysis of privacy policies and regulations, visualization of information collection practices, and design of a web-based interface for analyzing extracted privacy statements to identify vague, misleading, or incomplete privacy statements.
Desired Technical Skills: Good programming skills overall, and experience in using Jupyter and/or R for data analysis.
Desired Course(s): Software engineering, computer science, and information science students. Note: students with diverse backgrounds, including in technical fields, social sciences and humanities are encouraged to apply.
Other Desired Qualifications: Experience with machine learning, natural language processing techniques, HCI design and web development. interest in usable privacy, critical analysis of privacy policies and privacy related regulation.
Contact Info: rhythm.lab@yorku.ca
Lab Website: https://www.yorku.ca/lassonde/privacy/
Position Type: NSERC Undergraduate Student Research Award (USRA)
Open Positions: 2
Project Description: Modern sociotechnical systems share and collect vast amounts of information. These systems violate users’ privacy by ignoring the context in which the information is shared and by implementing privacy models that fail to incorporate contextual information norms.
Using techniques in natural language processing, machine learning, network, and data analysis, this project is set to explore the privacy implications of mobile apps, online platforms, and other systems in different social contexts/settings.
To tackle this challenge, the project will operationalize a cutting-edge privacy theory and methodologies to conduct an analysis of existing technologies and design privacy-enhancing tools.
Duties and Responsibilities: Students will develop privacy-enhancing mechanisms that ensure that information flows in accordance with users’ expectations and established societal norms and design. Specific tasks include: comprehensive literature review of existing methodologies and tools, analysis of privacy policies and regulations, visualization of information collection practices, and design of a web-based interface for analyzing extracted privacy statements to identify vague, misleading, or incomplete privacy statements.
Desired Technical Skills: Good programming skills overall, and experience in using Jupyter and/or R for data analysis.
Desired Course(s): Software engineering, computer science, and information science students. Note: students with diverse backgrounds, including in technical fields, social sciences and humanities are encouraged to apply.
Other Desired Qualifications: Experience with machine learning, natural language processing techniques, HCI design and web development. interest in usable privacy, critical analysis of privacy policies and privacy related regulation.
Professor: James Elder
Contact Info: jelder@yorku.ca
Lab Website: https://www.elderlab.yorku.ca/
Position Type: Lassonde Undergraduate Research Award (LURA); NSERC Undergraduate Student Research Award (USRA)
Open Positions: 1
Project Description: The local shape of the occluding contour of an object is known to constrain the local shape of the object surface, however these constraints are qualitative. While strict quantitative constraints relating the occluding contour to solid shapes are unlikely, we posit here that typical regularities of common objects and rules of projection induce dependencies that can be used to derive statistical estimates of quantitative solid shapes from the occluding contour. To explore this conjecture, we partition the problem into two parts: estimation of the 3D rim from the 2D occluding contour, and estimation of the visible surface shape from the estimated 3D rim. We train and evaluate statistical models on two distinct 3D object datasets and evaluate their ability to capture statistical regularities that enable 3D estimation of the object shape.
Line and ellipses are invariant under projection, making them convenient contour representations for the estimation of the 3D rim from the occluding contour. In this project we will therefore focus specifically on “elliptilinear” representations of the occluding contour and rim, i.e. piecewise elliptical curves, with linear intervals occurring with non-vanishing probability.
Duties and Responsibilities: The student will inherit 3D object datasets and software designed to recover optimal ellipti-linear representations of occluding contours. The student will validate the software and then analyze how these occluding contour representations relate to ellipti-linear approximations of the rim. Based on this analysis, the student will develop an algorithm to estimate the 3D shape of the rim from the 2D shape of the occluding contour. The student will have regular meetings with collaborator Yiming Qian (A*Star Singapore) as well as tri-weekly meetings with principal investigator Prof. James Elder.
At the end of the summer, the student will deliver code and labeled data in the form of a GitHub repository and an engineering report in LaTeX that documents the objectives, methods, results, and conclusions of this project and will present results at the Lassonde Annual Undergraduate Summer Student Research Conference.
Desired Technical Skills: MATLAB, and an aptitude in mathematics and statistics.
Desired Course(s): N/A.
Other Desired Qualifications: MATLAB, and an aptitude in mathematics and statistics.
Contact Info: jelder@yorku.ca
Lab Website: https://www.elderlab.yorku.ca/
Position Type: Lassonde Undergraduate Research Award (LURA); NSERC Undergraduate Student Research Award (USRA)
Open Positions: 1
Project Description: The local shape of the occluding contour of an object is known to constrain the local shape of the object surface, however these constraints are qualitative. While strict quantitative constraints relating the occluding contour to solid shapes are unlikely, we posit here that typical regularities of common objects and rules of projection induce dependencies that can be used to derive statistical estimates of quantitative solid shapes from the occluding contour. To explore this conjecture, we partition the problem into two parts: estimation of the 3D rim from the 2D occluding contour, and estimation of the visible surface shape from the estimated 3D rim. We train and evaluate statistical models on two distinct 3D object datasets and evaluate their ability to capture statistical regularities that enable 3D estimation of the object shape.
Line and ellipses are invariant under projection, making them convenient contour representations for the estimation of the 3D rim from the occluding contour. In this project we will therefore focus specifically on “elliptilinear” representations of the occluding contour and rim, i.e. piecewise elliptical curves, with linear intervals occurring with non-vanishing probability.
Duties and Responsibilities: The student will inherit 3D object datasets and software designed to recover optimal ellipti-linear representations of occluding contours. The student will validate the software and then analyze how these occluding contour representations relate to ellipti-linear approximations of the rim. Based on this analysis, the student will develop an algorithm to estimate the 3D shape of the rim from the 2D shape of the occluding contour. The student will have regular meetings with collaborator Yiming Qian (A*Star Singapore) as well as tri-weekly meetings with principal investigator Prof. James Elder.
At the end of the summer, the student will deliver code and labeled data in the form of a GitHub repository and an engineering report in LaTeX that documents the objectives, methods, results, and conclusions of this project and will present results at the Lassonde Annual Undergraduate Summer Student Research Conference.
Desired Technical Skills: MATLAB, and an aptitude in mathematics and statistics.
Desired Course(s): N/A.
Other Desired Qualifications: MATLAB, and an aptitude in mathematics and statistics.
Professor: James Elder
Contact Info: akajor@yorku.ca
Lab Website: https://www.elderlab.yorku.ca/
Position Type: Lassonde Undergraduate Research Award (LURA); NSERC Undergraduate Student Research Award (USRA)
Open Positions: 1
Project Description: The goal of the project is to research and develop computer vision algorithms, software, and specialized hardware for the analysis of mixed traffic at intersections. Multiple cameras will be employed. Road users will be detected and classified as motor vehicles, pedestrians, and bicycles. Motor vehicles will be further classified as vehicles with two wheels (motorcycles) and vehicles with 4 or more wheels (cars, trucks, buses). Road users will be geo-located within a 3D model of the intersection, tracked, and classified according to trajectory.
The research will include the design and development of systems for traffic counting and traffic anomaly detection. A system for 3D visualization of recorded or streaming traffic data (digital intersection) will also be designed and developed.
Supervisors: Sajjad Savoji, Bardia Esmaeili.
Duties and Responsibilities: Assist in ground-truthing and evaluation of algorithms for detection, classification, tracking, and trajectory classification of motor vehicles at intersections. Tabulate and analyze results, identifying failure modes. The student will have daily meetings with Master’s student Sajjad Savoji and PhD student Bardia Esmaeili to discuss progress, as well as tri-weekly meetings with principal investigator Prof. James Elder.
At the end of the summer the student will deliver code and labelled data in the form of a GitHub repository, as well as an engineering report in LaTeX that documents the objectives, methods, results, and conclusions of this project, and will present results at the Lassonde Annual Undergraduate Summer Student Research Conference.
Desired Technical Skills: Software – Python, MATLAB, and concepts – Familiarity with computer vision and 3D geometry skills preferred.
Desired Course(s): N/A.
Other Desired Qualifications: Software – Python, MATLAB, and concepts – Familiarity with computer vision and 3D geometry skills preferred.
Contact Info: akajor@yorku.ca
Lab Website: https://www.elderlab.yorku.ca/
Position Type: Lassonde Undergraduate Research Award (LURA); NSERC Undergraduate Student Research Award (USRA)
Open Positions: 1
Project Description: The goal of the project is to research and develop computer vision algorithms, software, and specialized hardware for the analysis of mixed traffic at intersections. Multiple cameras will be employed. Road users will be detected and classified as motor vehicles, pedestrians, and bicycles. Motor vehicles will be further classified as vehicles with two wheels (motorcycles) and vehicles with 4 or more wheels (cars, trucks, buses). Road users will be geo-located within a 3D model of the intersection, tracked, and classified according to trajectory.
The research will include the design and development of systems for traffic counting and traffic anomaly detection. A system for 3D visualization of recorded or streaming traffic data (digital intersection) will also be designed and developed.
Supervisors: Sajjad Savoji, Bardia Esmaeili.
Duties and Responsibilities: Assist in ground-truthing and evaluation of algorithms for detection, classification, tracking, and trajectory classification of motor vehicles at intersections. Tabulate and analyze results, identifying failure modes. The student will have daily meetings with Master’s student Sajjad Savoji and PhD student Bardia Esmaeili to discuss progress, as well as tri-weekly meetings with principal investigator Prof. James Elder.
At the end of the summer the student will deliver code and labelled data in the form of a GitHub repository, as well as an engineering report in LaTeX that documents the objectives, methods, results, and conclusions of this project, and will present results at the Lassonde Annual Undergraduate Summer Student Research Conference.
Desired Technical Skills: Software – Python, MATLAB, and concepts – Familiarity with computer vision and 3D geometry skills preferred.
Desired Course(s): N/A.
Other Desired Qualifications: Software – Python, MATLAB, and concepts – Familiarity with computer vision and 3D geometry skills preferred.
Professor: James Elder
Contact Info: akajor@yorku.ca
Lab Website: https://www.elderlab.yorku.ca/
Position Type: Lassonde Undergraduate Research Award (LURA); NSERC Undergraduate Student Research Award (USRA)
Open Positions: 1
Project Description: To be effective, a social robot must reliably detect and recognize people in all visual directions and in both near and far fields. A major challenge is the resolution/field-of-view trade-off, for which we have developed a novel attentive sensing solution. Panoramic low-resolution pre-attentive sensing is provided by an array of wide-angle cameras, while attentive sensing is achieved with a high-resolution, narrow field-of-view camera and a mirror-based gaze deflection system. Quantitative evaluation on a novel dataset shows that this attentive sensing strategy can yield good panoramic face recognition performance in the wild out to distances of ~35m.
Currently the system operates at about 0.3 fixations per second, far slower than the human fixation rate of 2-3 fixations per second. In this project we will redesign our system to come closer to human performance. This will involve lightening the payload, updating the motor, improving the control system, and updating the attentive camera to allow manual control of focus so that the camera can change focus while making a saccade.
Supervisors: Mohammad Akhavan, Helio Perroni-Filho.
Duties and Responsibilities: The student will work closely with the supervisors to develop and test algorithms for video-based highway traffic analytics. The student will have daily meetings with graduate student Mohammad Akhavan and Senior Robotics Engineer Helio Perroni-Filho to discuss progress, as well as tri-weekly meetings with principal investigator Prof. James Elder.
At the end of the summer the student will demonstrate the redesigned attentive sensor system, deliver documented software in the form of a GitHub repository and deliver an engineering report in LaTeX that documents the objectives, methods, results and conclusions of this project, and will present results at the Lassonde Annual Undergraduate Summer Student Research Conference.
Desired Technical Skills: Software skills, control theory and algorithms, systems design.
Desired Course(s): N/A.
Other Desired Qualifications: Software skills, control theory and algorithms, systems design.
Contact Info: akajor@yorku.ca
Lab Website: https://www.elderlab.yorku.ca/
Position Type: Lassonde Undergraduate Research Award (LURA); NSERC Undergraduate Student Research Award (USRA)
Open Positions: 1
Project Description: To be effective, a social robot must reliably detect and recognize people in all visual directions and in both near and far fields. A major challenge is the resolution/field-of-view trade-off, for which we have developed a novel attentive sensing solution. Panoramic low-resolution pre-attentive sensing is provided by an array of wide-angle cameras, while attentive sensing is achieved with a high-resolution, narrow field-of-view camera and a mirror-based gaze deflection system. Quantitative evaluation on a novel dataset shows that this attentive sensing strategy can yield good panoramic face recognition performance in the wild out to distances of ~35m.
Currently the system operates at about 0.3 fixations per second, far slower than the human fixation rate of 2-3 fixations per second. In this project we will redesign our system to come closer to human performance. This will involve lightening the payload, updating the motor, improving the control system, and updating the attentive camera to allow manual control of focus so that the camera can change focus while making a saccade.
Supervisors: Mohammad Akhavan, Helio Perroni-Filho.
Duties and Responsibilities: The student will work closely with the supervisors to develop and test algorithms for video-based highway traffic analytics. The student will have daily meetings with graduate student Mohammad Akhavan and Senior Robotics Engineer Helio Perroni-Filho to discuss progress, as well as tri-weekly meetings with principal investigator Prof. James Elder.
At the end of the summer the student will demonstrate the redesigned attentive sensor system, deliver documented software in the form of a GitHub repository and deliver an engineering report in LaTeX that documents the objectives, methods, results and conclusions of this project, and will present results at the Lassonde Annual Undergraduate Summer Student Research Conference.
Desired Technical Skills: Software skills, control theory and algorithms, systems design.
Desired Course(s): N/A.
Other Desired Qualifications: Software skills, control theory and algorithms, systems design.
Professor: James Elder
Contact Info: jelder@yorku.ca
Lab Website: https://www.elderlab.yorku.ca/
Position Type: Lassonde Undergraduate Research Award (LURA); NSERC Undergraduate Student Research Award (USRA)
Open Positions: 1
Project Description: Our pilot studies suggest that both humans and networks rely strongly on elevation in the image as a cue to depth, consistent with the dominance of ground plane surfaces. Moreover, our recent work on scene classification has suggested that human awareness of scene semantics can precede and inform awareness of 3D spatial layout. These findings motivate a novel approach to monocular depth estimation in which depth is inferred directly from semantic surface labels and the statistics of 3D spatial relationships between surfaces.
To explore this novel approach, we will again employ a dataset of planar projections generated from the SYNS dataset. State-of-the-art deep semantic and instance segmentation systems will be applied to estimate the semantic surface labels and individuate objects. In the first stage of inference, known view geometry will be used to assign depths to all ground-plane semantic categories (e.g. road, sidewalk, field). In a second stage of inference, semantic regions adjacent to ground-plane regions will be assigned depths based upon vertical propagation from assigned ground-plane depths. In a third stage, for regions not adjacent to ground-plane regions, depths will be estimated based upon adjacent regions already assigned depths and a learned table of expected relative depth for pairs of adjacent semantic categories. Our goal is to carefully refine the algorithm by also incorporating instance segmentation and more carefully encoding 3D spatial relationships between semantic categories. Results will be compared against state-of-the-art deep network systems.
Supervisors: Alek Trajcevski, Hossein Hosseini.
Duties and Responsibilities: The student will work closely with the supervisors to develop and test algorithms for monocular estimation of absolute and relative depth on held-out datasets. The student will have daily meetings with graduate students Alek Trajcevski and Hossein Hosseini to discuss progress, as well as tri-weekly meetings with principal investigator Prof. James Elder.
At the end of the summer the student will deliver documented software in the form of a GitHub repository and an engineering report in LaTeX that documents the objectives, methods, results, and conclusions of this project, and will present results at the Lassonde Annual Undergraduate Summer Student Research Conference.
Desired Technical Skills: Software skills, 3D geometry, machine learning.
Desired Course(s): N/A.
Other Desired Qualifications: Software skills, 3D geometry, machine learning.
Contact Info: jelder@yorku.ca
Lab Website: https://www.elderlab.yorku.ca/
Position Type: Lassonde Undergraduate Research Award (LURA); NSERC Undergraduate Student Research Award (USRA)
Open Positions: 1
Project Description: Our pilot studies suggest that both humans and networks rely strongly on elevation in the image as a cue to depth, consistent with the dominance of ground plane surfaces. Moreover, our recent work on scene classification has suggested that human awareness of scene semantics can precede and inform awareness of 3D spatial layout. These findings motivate a novel approach to monocular depth estimation in which depth is inferred directly from semantic surface labels and the statistics of 3D spatial relationships between surfaces.
To explore this novel approach, we will again employ a dataset of planar projections generated from the SYNS dataset. State-of-the-art deep semantic and instance segmentation systems will be applied to estimate the semantic surface labels and individuate objects. In the first stage of inference, known view geometry will be used to assign depths to all ground-plane semantic categories (e.g. road, sidewalk, field). In a second stage of inference, semantic regions adjacent to ground-plane regions will be assigned depths based upon vertical propagation from assigned ground-plane depths. In a third stage, for regions not adjacent to ground-plane regions, depths will be estimated based upon adjacent regions already assigned depths and a learned table of expected relative depth for pairs of adjacent semantic categories. Our goal is to carefully refine the algorithm by also incorporating instance segmentation and more carefully encoding 3D spatial relationships between semantic categories. Results will be compared against state-of-the-art deep network systems.
Supervisors: Alek Trajcevski, Hossein Hosseini.
Duties and Responsibilities: The student will work closely with the supervisors to develop and test algorithms for monocular estimation of absolute and relative depth on held-out datasets. The student will have daily meetings with graduate students Alek Trajcevski and Hossein Hosseini to discuss progress, as well as tri-weekly meetings with principal investigator Prof. James Elder.
At the end of the summer the student will deliver documented software in the form of a GitHub repository and an engineering report in LaTeX that documents the objectives, methods, results, and conclusions of this project, and will present results at the Lassonde Annual Undergraduate Summer Student Research Conference.
Desired Technical Skills: Software skills, 3D geometry, machine learning.
Desired Course(s): N/A.
Other Desired Qualifications: Software skills, 3D geometry, machine learning.
Professor: James Elder
Contact Info: akajor@yorku.ca
Lab Website: https://www.elderlab.yorku.ca/
Position Type: Lassonde Undergraduate Research Award (LURA); NSERC Undergraduate Student Research Award (USRA)
Open Positions: 1
Project Description: Traffic congestion is a major challenge throughout the world. It affects commute time, mobility, and accessibility and is a driving factor in increasing the harmful gasses that are the main culprits of the greenhouse effect. Congestion can potentially be mitigated over short time scales through improvements to signal timing and rapid detection and resolution of traffic incidents, and over longer time scales through strategic roadway improvements and optimization of public transit systems. However, all of these mitigations depend critically on an accurate understanding of lane-by-lane traffic density and speed distributions. Historically, these data sets are obtained from embedded inductive loops, but these are expensive to maintain. This motivates the development of fully automated AI-based video analytics systems to accurately measure traffic density and speed and to detect anomalous traffic incidents.
In the proposed research we will focus specifically on highways, which in large metropolitan areas are typically the key pinch points that determine overall congestion. One challenge to achieving our objective is the volume of video data required to densely sample these highways and to react to measurements in a timely way. Distributing this quantity of data to the cloud or a central control centre for processing entails an enormous investment in communication and storage and, despite advances in low-latency communication protocols, will inevitably lead to delays in response. It also raises privacy and data security issues. These challenges motivate an edge solution, where video is processed by embedded systems located on-site, and only low-bandwidth and anonymized derived analytics products are communicated centrally. In the proposed research, we plan to design such an embedded system, optimized for highway deployment.
The system will be designed for mounting at 10m height on either meridian or side poles, and will incorporate multiple cameras that together provide a field of view covering all lanes of traffic in both directions. These will be paired with an advanced embedded AI device such as an NVIDIA Jetson AGX Orin. Deep networks will be pruned and quantized (INT8, FP16, Mixed Precision) to meet throughput requirements. Provided functionality will include: Automatic camera calibration and highway understanding that allows events localized in the image to be precisely back projected to highway ground coordinates, detection, classification, and segmentation of motor vehicles, vehicle speed estimation, detection of anomalies, including accidents and stopped vehicles, reporting of anomalous vehicles through automatic number plate recognition.
Supervisor: Shreejal Trevedi.
Duties and Responsibilities: The student will work closely with the supervisors to develop and test algorithms for video-based highway traffic analytics. The student will have daily meetings with graduate student Shreejal Trivedi to discuss progress, as well as tri-weekly meetings with principal investigator Prof. James Elder and with project partners at the Ministry of Transportation Ontario.
The planned outcome of this research project is a prototype embedded traffic analytics products optimized for highway deployment. We hope to ultimately commercialize this device through licensing or a start-up company. We believe this device will ultimately replace costly inductive loop systems and contribute to better traffic analytics, thus ultimately reducing congestion and greenhouse gas emissions.
At the end of the summer the student will deliver documented software in the form of a GitHub repository and an engineering report in LaTeX that documents the objectives, methods, results, and conclusions of this project, and will present results at the Lassonde Annual Undergraduate Summer Student Research Conference.
Desired Technical Skills: Software skills, and systems hardware skills.
Desired Course(s): N/A.
Other Desired Qualifications: Software skills, and systems hardware skills.
Contact Info: akajor@yorku.ca
Lab Website: https://www.elderlab.yorku.ca/
Position Type: Lassonde Undergraduate Research Award (LURA); NSERC Undergraduate Student Research Award (USRA)
Open Positions: 1
Project Description: Traffic congestion is a major challenge throughout the world. It affects commute time, mobility, and accessibility and is a driving factor in increasing the harmful gasses that are the main culprits of the greenhouse effect. Congestion can potentially be mitigated over short time scales through improvements to signal timing and rapid detection and resolution of traffic incidents, and over longer time scales through strategic roadway improvements and optimization of public transit systems. However, all of these mitigations depend critically on an accurate understanding of lane-by-lane traffic density and speed distributions. Historically, these data sets are obtained from embedded inductive loops, but these are expensive to maintain. This motivates the development of fully automated AI-based video analytics systems to accurately measure traffic density and speed and to detect anomalous traffic incidents.
In the proposed research we will focus specifically on highways, which in large metropolitan areas are typically the key pinch points that determine overall congestion. One challenge to achieving our objective is the volume of video data required to densely sample these highways and to react to measurements in a timely way. Distributing this quantity of data to the cloud or a central control centre for processing entails an enormous investment in communication and storage and, despite advances in low-latency communication protocols, will inevitably lead to delays in response. It also raises privacy and data security issues. These challenges motivate an edge solution, where video is processed by embedded systems located on-site, and only low-bandwidth and anonymized derived analytics products are communicated centrally. In the proposed research, we plan to design such an embedded system, optimized for highway deployment.
The system will be designed for mounting at 10m height on either meridian or side poles, and will incorporate multiple cameras that together provide a field of view covering all lanes of traffic in both directions. These will be paired with an advanced embedded AI device such as an NVIDIA Jetson AGX Orin. Deep networks will be pruned and quantized (INT8, FP16, Mixed Precision) to meet throughput requirements. Provided functionality will include: Automatic camera calibration and highway understanding that allows events localized in the image to be precisely back projected to highway ground coordinates, detection, classification, and segmentation of motor vehicles, vehicle speed estimation, detection of anomalies, including accidents and stopped vehicles, reporting of anomalous vehicles through automatic number plate recognition.
Supervisor: Shreejal Trevedi.
Duties and Responsibilities: The student will work closely with the supervisors to develop and test algorithms for video-based highway traffic analytics. The student will have daily meetings with graduate student Shreejal Trivedi to discuss progress, as well as tri-weekly meetings with principal investigator Prof. James Elder and with project partners at the Ministry of Transportation Ontario.
The planned outcome of this research project is a prototype embedded traffic analytics products optimized for highway deployment. We hope to ultimately commercialize this device through licensing or a start-up company. We believe this device will ultimately replace costly inductive loop systems and contribute to better traffic analytics, thus ultimately reducing congestion and greenhouse gas emissions.
At the end of the summer the student will deliver documented software in the form of a GitHub repository and an engineering report in LaTeX that documents the objectives, methods, results, and conclusions of this project, and will present results at the Lassonde Annual Undergraduate Summer Student Research Conference.
Desired Technical Skills: Software skills, and systems hardware skills.
Desired Course(s): N/A.
Other Desired Qualifications: Software skills, and systems hardware skills.
Professor: James Elder
Contact Info: akajor@yorku.ca
Lab Website: https://www.elderlab.yorku.ca/
Position Type: Lassonde Undergraduate Research Award (LURA); NSERC Undergraduate Student Research Award (USRA)
Open Positions: 1
Project Description: ImageNet-trained networks fail to develop a configural representation of object shape. A main reason for this failure may be the impoverished nature of the image labelling task. While the human brain of course supports object recognition, humans form much richer representations of objects that allow us to appreciate many diverse qualities of objects and their spatial relationships to each other and the rest of the visual scene. To support this diverse experience of objects, we hypothesize that the brain establishes useful intermediate representations that can subserve multiple tasks. In this project, we explore this hypothesis computationally by modifying existing deep network architectures to expand the task from object labelling to also include object localization, segmentation, and monocular depth estimation. Critically, losses will be applied not only at the outputs of each respective head, but also at intermediate stages of computation. These intermediate losses will take the form of regional forms of their respective tasks, with spatial extent matched to the nominal receptive field size.
We hypothesize that this will lead to more holistic processing as the network takes advantage of synergies between the tasks, and local shortcuts that might be useful for recognition are disfavoured since they do not contribute to other tasks.
Supervisor: Nima Vahdat.
Duties and Responsibilities: The student will work closely with the supervisors to develop and test algorithms for video-based highway traffic analytics. The student will have daily meetings with graduate student Nima Vahdat to discuss progress, as well as tri-weekly meetings with principal investigator Prof. James Elder.
At the end of the summer the student will deliver documented software in the form of a GitHub repository and an engineering report in LaTeX that documents the objectives, methods, results, and conclusions of this project, and will present results at the Lassonde Annual Undergraduate Summer Student Research Conference.
Desired Technical Skills: Software skills, Python, PyTorch, and deep learning.
Desired Course(s): N/A.
Other Desired Qualifications: Software skills, Python, PyTorch, and deep learning.
Contact Info: akajor@yorku.ca
Lab Website: https://www.elderlab.yorku.ca/
Position Type: Lassonde Undergraduate Research Award (LURA); NSERC Undergraduate Student Research Award (USRA)
Open Positions: 1
Project Description: ImageNet-trained networks fail to develop a configural representation of object shape. A main reason for this failure may be the impoverished nature of the image labelling task. While the human brain of course supports object recognition, humans form much richer representations of objects that allow us to appreciate many diverse qualities of objects and their spatial relationships to each other and the rest of the visual scene. To support this diverse experience of objects, we hypothesize that the brain establishes useful intermediate representations that can subserve multiple tasks. In this project, we explore this hypothesis computationally by modifying existing deep network architectures to expand the task from object labelling to also include object localization, segmentation, and monocular depth estimation. Critically, losses will be applied not only at the outputs of each respective head, but also at intermediate stages of computation. These intermediate losses will take the form of regional forms of their respective tasks, with spatial extent matched to the nominal receptive field size.
We hypothesize that this will lead to more holistic processing as the network takes advantage of synergies between the tasks, and local shortcuts that might be useful for recognition are disfavoured since they do not contribute to other tasks.
Supervisor: Nima Vahdat.
Duties and Responsibilities: The student will work closely with the supervisors to develop and test algorithms for video-based highway traffic analytics. The student will have daily meetings with graduate student Nima Vahdat to discuss progress, as well as tri-weekly meetings with principal investigator Prof. James Elder.
At the end of the summer the student will deliver documented software in the form of a GitHub repository and an engineering report in LaTeX that documents the objectives, methods, results, and conclusions of this project, and will present results at the Lassonde Annual Undergraduate Summer Student Research Conference.
Desired Technical Skills: Software skills, Python, PyTorch, and deep learning.
Desired Course(s): N/A.
Other Desired Qualifications: Software skills, Python, PyTorch, and deep learning.
Professor: James Elder
Contact Info: akajor@yorku.ca
Lab Website: https://www.elderlab.yorku.ca/
Position Type: Lassonde Undergraduate Research Award (LURA); NSERC Undergraduate Student Research Award (USRA)
Open Positions: 1
Project Description: We have developed an attentive social robot that is capable of attentively searching for specific individuals, recognizing them at a distance using facial recognition software, and approaching them for more information. However, these behaviours have only been demonstrated under control conditions and for short operating periods. In this project, we will develop the enhancements to software, hardware, and protocol required to allow the robot to operate continuously and reliably for long durations, collecting useful social information in a multi-room environment.
Supervisor: Helio Perroni-Filho.
Duties and Responsibilities: The student will work closely with the supervisors to develop and test algorithms for video-based highway traffic analytics. The student will have daily meetings with Senior Robotics Engineer Helio Perroni-Filho to discuss progress, as well as tri-weekly meetings with principal investigator Prof. James Elder.
At the end of the summer the student will demonstrate long-term operation of the robot and deliver an engineering report in LaTeX that documents the objectives, methods, results, and conclusions of this project, and will present results at the Lassonde Annual Undergraduate Summer Student Research Conference.
Desired Technical Skills: Software skills, and systems design.
Desired Course(s): N/A.
Other Desired Qualifications: Software skills, and systems design.
Contact Info: akajor@yorku.ca
Lab Website: https://www.elderlab.yorku.ca/
Position Type: Lassonde Undergraduate Research Award (LURA); NSERC Undergraduate Student Research Award (USRA)
Open Positions: 1
Project Description: We have developed an attentive social robot that is capable of attentively searching for specific individuals, recognizing them at a distance using facial recognition software, and approaching them for more information. However, these behaviours have only been demonstrated under control conditions and for short operating periods. In this project, we will develop the enhancements to software, hardware, and protocol required to allow the robot to operate continuously and reliably for long durations, collecting useful social information in a multi-room environment.
Supervisor: Helio Perroni-Filho.
Duties and Responsibilities: The student will work closely with the supervisors to develop and test algorithms for video-based highway traffic analytics. The student will have daily meetings with Senior Robotics Engineer Helio Perroni-Filho to discuss progress, as well as tri-weekly meetings with principal investigator Prof. James Elder.
At the end of the summer the student will demonstrate long-term operation of the robot and deliver an engineering report in LaTeX that documents the objectives, methods, results, and conclusions of this project, and will present results at the Lassonde Annual Undergraduate Summer Student Research Conference.
Desired Technical Skills: Software skills, and systems design.
Desired Course(s): N/A.
Other Desired Qualifications: Software skills, and systems design.
Professor: James Elder
Contact Info: akajor@yorku.ca
Lab Website: https://www.elderlab.yorku.ca/
Position Type: Lassonde Undergraduate Research Award (LURA); NSERC Undergraduate Student Research Award (USRA)
Open Positions: 1
Project Description: We have developed an attentive social robot that is capable of attentively searching for specific individuals, recognizing them at a distance using facial recognition software, and approaching them for more information. One limitation of our attentive sensor is that the field of view is partially occluded in four directions by the aluminum posts that support the motorized mirror. Fortunately, due to focal blur the occlusion is only partial, which suggests that it could potentially be calibrated out. The goal of this project is to develop and evaluate algorithms for minimizing this artifact.
Supervisor: Helio Perroni-Filho.
Duties and Responsibilities: The student will work closely with the supervisors to develop and test algorithms for removing the occlusion artifact. The student will have daily meetings with Senior Robotics Engineer Helio Perroni-Filho to discuss progress, as well as tri-weekly meetings with principal investigator Prof. James Elder.
At the end of the summer the student will demonstrate the occlusion removal method as implemented in the robot and deliver an engineering report in LaTeX that documents the objectives, methods, results, and conclusions of this project, and will present results at the Lassonde Annual Undergraduate Summer Student Research Conference.
Desired Technical Skills: Software skills, and systems design.
Desired Course(s): N/A.
Other Desired Qualifications: Software skills, and systems design.
Contact Info: akajor@yorku.ca
Lab Website: https://www.elderlab.yorku.ca/
Position Type: Lassonde Undergraduate Research Award (LURA); NSERC Undergraduate Student Research Award (USRA)
Open Positions: 1
Project Description: We have developed an attentive social robot that is capable of attentively searching for specific individuals, recognizing them at a distance using facial recognition software, and approaching them for more information. One limitation of our attentive sensor is that the field of view is partially occluded in four directions by the aluminum posts that support the motorized mirror. Fortunately, due to focal blur the occlusion is only partial, which suggests that it could potentially be calibrated out. The goal of this project is to develop and evaluate algorithms for minimizing this artifact.
Supervisor: Helio Perroni-Filho.
Duties and Responsibilities: The student will work closely with the supervisors to develop and test algorithms for removing the occlusion artifact. The student will have daily meetings with Senior Robotics Engineer Helio Perroni-Filho to discuss progress, as well as tri-weekly meetings with principal investigator Prof. James Elder.
At the end of the summer the student will demonstrate the occlusion removal method as implemented in the robot and deliver an engineering report in LaTeX that documents the objectives, methods, results, and conclusions of this project, and will present results at the Lassonde Annual Undergraduate Summer Student Research Conference.
Desired Technical Skills: Software skills, and systems design.
Desired Course(s): N/A.
Other Desired Qualifications: Software skills, and systems design.
Professor: James Elder
Contact Info: akajor@yorku.ca
Lab Website: https://www.elderlab.yorku.ca/
Position Type: Lassonde Undergraduate Research Award (LURA); NSERC Undergraduate Student Research Award (USRA)
Open Positions: 1
Project Description: We have developed a novel Markov Chain Monte Carlo method for created 2D planar shapes with prescribed curvature statistics. These stimuli are useful for behavioural and neurophysiological experiments in human and animal models, and to evaluate computational models for object perception.
Unfortunately, our software contains some bugs; occasionally, the bounding contours of these shapes self-intersect, leading to pathological stimuli. The goal of this project is to find and correct these bugs.
Duties and Responsibilities: The student will work closely with the supervisor to debug this software and deliver a corrected version in the form of a GitHub repository. The student will have regular meetings with principal investigator Prof. James Elder.
At the end of the summer the student will demonstrate and validate the corrected software and deliver an engineering report in LaTeX that documents the objectives, methods, results, and conclusions of this project, and will present results at the Lassonde Annual Undergraduate Summer Student Research Conference.
Desired Technical Skills: Software skills, mathematical skills, and statistics.
Desired Course(s): N/A.
Other Desired Qualifications: Software skills, mathematical skills, and statistics.
Contact Info: akajor@yorku.ca
Lab Website: https://www.elderlab.yorku.ca/
Position Type: Lassonde Undergraduate Research Award (LURA); NSERC Undergraduate Student Research Award (USRA)
Open Positions: 1
Project Description: We have developed a novel Markov Chain Monte Carlo method for created 2D planar shapes with prescribed curvature statistics. These stimuli are useful for behavioural and neurophysiological experiments in human and animal models, and to evaluate computational models for object perception.
Unfortunately, our software contains some bugs; occasionally, the bounding contours of these shapes self-intersect, leading to pathological stimuli. The goal of this project is to find and correct these bugs.
Duties and Responsibilities: The student will work closely with the supervisor to debug this software and deliver a corrected version in the form of a GitHub repository. The student will have regular meetings with principal investigator Prof. James Elder.
At the end of the summer the student will demonstrate and validate the corrected software and deliver an engineering report in LaTeX that documents the objectives, methods, results, and conclusions of this project, and will present results at the Lassonde Annual Undergraduate Summer Student Research Conference.
Desired Technical Skills: Software skills, mathematical skills, and statistics.
Desired Course(s): N/A.
Other Desired Qualifications: Software skills, mathematical skills, and statistics.
Professor: Hina Tabassum
Contact Info: hinat@yorku.ca
Lab Website: https://sites.google.com/a/kaust.edu.sa/hina-tabassum/
Position Type: Lassonde Undergraduate Research Award (LURA); NSERC Undergraduate Student Research Award (USRA)
Open Positions: 1
Project Description: Wi-Fi access points are evolving from mere communication devices to sensing instruments, leveraging Channel State Information (CSI) extraction capabilities. CSI captures variations in Wi-Fi or Radio Frequency (RF) signals as they move through a physical space, interacting with objects or human bodies, causing phenomena like reflection, diffraction, and scattering. These interactions result in multipath effects, which convey valuable information about the surrounding environment, including human movements, locations, and the state of objects. Wi-Fi sensing is a low-cost privacy-preserving solution that benefits a variety of industries working on remote health monitoring, assisted-living, activity recognition, exceptional event identification, risk assessment and alert generation. This project will focus on CSI human activity data collection using Raspberry-Pi and develop non-contrastive self-supervised machine learning models to predict the accuracy of human activity recognition. A comparative analysis will be performed among contrastive and non-contrastive deep learning models by considering different augmentations of the CSI image.
Duties and Responsibilities: Collection of channel state information (CSI) using Raspberry-Pi. Explore the feasibility and performance of contrastive and non-contrastive self-supervised learning methods.
Desired Technical Skills: Programming skills (Python, MATLAB) – Familiarity with wireless communications and deep learning concepts preferred. Prior experience with Raspberry-Pi will be an asset.
Desired Course(s): Courses related to Basics of Machine learning, Optimization theory, and Wireless communications would be preferred.
Other Desired Qualifications: N/A.
Contact Info: hinat@yorku.ca
Lab Website: https://sites.google.com/a/kaust.edu.sa/hina-tabassum/
Position Type: Lassonde Undergraduate Research Award (LURA); NSERC Undergraduate Student Research Award (USRA)
Open Positions: 1
Project Description: Wi-Fi access points are evolving from mere communication devices to sensing instruments, leveraging Channel State Information (CSI) extraction capabilities. CSI captures variations in Wi-Fi or Radio Frequency (RF) signals as they move through a physical space, interacting with objects or human bodies, causing phenomena like reflection, diffraction, and scattering. These interactions result in multipath effects, which convey valuable information about the surrounding environment, including human movements, locations, and the state of objects. Wi-Fi sensing is a low-cost privacy-preserving solution that benefits a variety of industries working on remote health monitoring, assisted-living, activity recognition, exceptional event identification, risk assessment and alert generation. This project will focus on CSI human activity data collection using Raspberry-Pi and develop non-contrastive self-supervised machine learning models to predict the accuracy of human activity recognition. A comparative analysis will be performed among contrastive and non-contrastive deep learning models by considering different augmentations of the CSI image.
Duties and Responsibilities: Collection of channel state information (CSI) using Raspberry-Pi. Explore the feasibility and performance of contrastive and non-contrastive self-supervised learning methods.
Desired Technical Skills: Programming skills (Python, MATLAB) – Familiarity with wireless communications and deep learning concepts preferred. Prior experience with Raspberry-Pi will be an asset.
Desired Course(s): Courses related to Basics of Machine learning, Optimization theory, and Wireless communications would be preferred.
Other Desired Qualifications: N/A.
Professor: Kiemute Oyibo
Contact Info: koyibo@yorku.ca
Lab Website: https://lassonde.yorku.ca/users/koyibo
Position Type: Lassonde Undergraduate Research Award (LURA); NSERC Undergraduate Student Research Award (USRA)
Open Positions: 2
Project Description: Many courses in Social Science, Health Science, and Computer Science that require memorization are becoming more and more challenging for many college and university students, especially with the ever-increasing volume of textbooks and course materials due to the growing body of knowledge. In courses, such as psychology, biology, and Human-Computer Interaction (HCI) that are memorization intensive, students are often overwhelmed by the lengthy course materials, readings, and the uphill challenge to retain and recall most of the content taught in class. Given the many challenges that most students face in higher education, including having to work to help pay for their university education, they often lack sufficient time to read and properly understand the taught material. Hence, there is a need to find effective ways to support student learning. We argue that serious games can be utilized to support student learning in memory-intensive courses to increase content comprehension and retention. Serious games are interactive applications that use game elements for education purposes rather than entertainment. They are increasingly being used in the education domain to support student learning. In this project, we aim to design, implement, and evaluate a mnemonic-based serious game to help students learn and retain taught material easily in memorization-intensive courses individually and collaboratively.
Duties and Responsibilities: Requirements gathering and analysis, prototyping, application programming and evaluation, data analysis, report writing and presentation.
Desired Technical Skills: Prototyping with tools such as Figma and programming on the mobile platform (e.g., Android, cross-platform).
Desired Course(s): Software Engineering, LE/EECS 4441 3.00 – Human-Computer Interaction, LE/EECS 3461 3.00 – User Interfaces
Other Desired Qualifications: Ability to work independently as well as in a team.
Contact Info: koyibo@yorku.ca
Lab Website: https://lassonde.yorku.ca/users/koyibo
Position Type: Lassonde Undergraduate Research Award (LURA); NSERC Undergraduate Student Research Award (USRA)
Open Positions: 2
Project Description: Many courses in Social Science, Health Science, and Computer Science that require memorization are becoming more and more challenging for many college and university students, especially with the ever-increasing volume of textbooks and course materials due to the growing body of knowledge. In courses, such as psychology, biology, and Human-Computer Interaction (HCI) that are memorization intensive, students are often overwhelmed by the lengthy course materials, readings, and the uphill challenge to retain and recall most of the content taught in class. Given the many challenges that most students face in higher education, including having to work to help pay for their university education, they often lack sufficient time to read and properly understand the taught material. Hence, there is a need to find effective ways to support student learning. We argue that serious games can be utilized to support student learning in memory-intensive courses to increase content comprehension and retention. Serious games are interactive applications that use game elements for education purposes rather than entertainment. They are increasingly being used in the education domain to support student learning. In this project, we aim to design, implement, and evaluate a mnemonic-based serious game to help students learn and retain taught material easily in memorization-intensive courses individually and collaboratively.
Duties and Responsibilities: Requirements gathering and analysis, prototyping, application programming and evaluation, data analysis, report writing and presentation.
Desired Technical Skills: Prototyping with tools such as Figma and programming on the mobile platform (e.g., Android, cross-platform).
Desired Course(s): Software Engineering, LE/EECS 4441 3.00 – Human-Computer Interaction, LE/EECS 3461 3.00 – User Interfaces
Other Desired Qualifications: Ability to work independently as well as in a team.
Professor: Razieh (Neda) Salahandish
Contact Info: raziehs@yorku.ca
Lab Website: https://lab-ha.eecs.yorku.ca/
Position Type: Lassonde Undergraduate Research Award (LURA); NSERC Undergraduate Student Research Award (USRA)
Open Positions: 3
Project Description: Alzheimer’s disease (AD), as a progressive neurodegenerative disorder involving impairment in memory, perception, and language abilities, possesses a high prevalence among elderlies aged 65 and higher (e.g. ~6.5 million Americans, 2022, and is projected to grow to 13 million by 2050), and is associated with excess morbidity and mortality. The disease is also known to be asymptomatic for years, in terms of cognitive decline or memory loss manifestations, while the biological cues are silently present and progressing. Early diagnosis of AD contributes significantly to disease management, impeding the progression into severe forms such as late-stage dementia.
Currently, several biomarkers are considered for screening AD, detection of some includes invasive methods, while others are based on neuroimaging and neuropsychological cognitive tasks, such as the Montreal Cognitive Assessment. While imaging and brain signal acquisition modalities demand centralized facilities, making them expensive and time-consuming, invasive sample collection procedures are prohibitory.
While the current non-invasive brain wave monitoring techniques, such as electroencephalograms (EEG) or brain imaging modalities, contribute significantly in depicting manifestations of disease, there still exists no gold standard clinical non-invasive solution for disease progression tracking. To this end, different eye segments can be evaluated for discovering AD-associated representations, including the retina, iris, or simply eye movement. Pathologies of AD in the neurosensory layers of the retina are evident, and hence, interest in the extraction of retinal Optical Coherence Tomography (OCT) features representative of the disease progression and in the early stages of cognitive impairment is growing. The biological components in retina tissue, compared to non-impaired individuals, provide further support for the utility of the retina in AD diagnosis. Consequently using OCT images for this purpose would be of paramount value, even in mild cases, upon discovery of the AD-specific features in the retina and optic nerve. The efforts to find such patterns distinguishing AD patients and the age-matched cohort, have been carried out since 2001, with instances such as retinal nerve fiber layer (rNFL) thinning in AD patients revealed. As another non-invasive approach, eye tracking is also of interest, since it serves as a language and culturally independent method for large-scale screening, and has been previously shown to be effective for distinguishing patients with neurological disorders like autism spectrum disorder (ASD). In this project, we first will conduct a clinical study to evaluate the clinical value and efficacy of eye-movement tracking, OCT features, and EEG signals in discovering patterns distinguishing AD patients from the control group. The data from this study will assist in determining the most discriminative AD markers in a first-time biomarker discovery initiative establishing correlations between markers, in particular with more established OCT features. In the second phase, we would undertake a “panel” approach, i.e. simultaneous detection of serval markers. “Smart goggles” will be developed with embedded cameras for eye tracking with extended electrodes on the temporal lobes for scalp EEG acquisition. The precise specifications of the final location of electrodes, their types, and hence the characteristics of the signals will be determined based on the outcomes of the first phase of the study. The data acquired from all these sensory modalities will be wirelessly communicated to a portable processor (e.g., a cellphone), and analyzed via Machine Learning (ML) algorithms for automated assessment of AD progression. This integrated wearable system will serve as a Point-of-Care (PoC) platform, enabling early detection of AD, and monitoring the effect of treatments on AD prognosis.
Duties and Responsibilities: Assist in acquiring OCT images, eye movement tracking, and brain signals from the recruited AD patients and age-matched healthy individuals. Assist in developing machine learning (ML) algorithms for feature extraction in the collected data and providing correlations and specific markers associated with AD. Assist in the design and development of smart goggles for recording image-based and temporal sensory information. Assist in functionality assessment and optimization of the developed smart goggles and a pilot study in a clinical cohort for evaluating the utility of the system in detecting AD cases.
Desired Technical Skills: Background in Electrical, Computer, and Software Engineering: Demonstrated expertise in the fundamentals and advanced concepts of these engineering disciplines, including systems design, programming, and hardware-software integration. Experience in AI Model Development: Skilled in training machine learning models, labeling datasets, and programming for AI applications. This includes hands-on experience in developing, supervising, and evaluating the performance of AI models across various applications. Experience in Circuit Design: Knowledge in designing circuits specifically for EEG electrodes, including aspects of miniaturization, signal integrity, and noise reduction. Programming Skills: Experience in programming languages commonly used in Electrical, Computer, and Software Engineering, such as Python, C++, Java, and MATLAB. Team Collaboration: Experience in working collaboratively in interdisciplinary teams that involve complex electrical, computer, and software engineering tasks.
Desired Course(s): N/A.
Other Desired Qualifications: N/A.
Contact Info: raziehs@yorku.ca
Lab Website: https://lab-ha.eecs.yorku.ca/
Position Type: Lassonde Undergraduate Research Award (LURA); NSERC Undergraduate Student Research Award (USRA)
Open Positions: 3
Project Description: Alzheimer’s disease (AD), as a progressive neurodegenerative disorder involving impairment in memory, perception, and language abilities, possesses a high prevalence among elderlies aged 65 and higher (e.g. ~6.5 million Americans, 2022, and is projected to grow to 13 million by 2050), and is associated with excess morbidity and mortality. The disease is also known to be asymptomatic for years, in terms of cognitive decline or memory loss manifestations, while the biological cues are silently present and progressing. Early diagnosis of AD contributes significantly to disease management, impeding the progression into severe forms such as late-stage dementia.
Currently, several biomarkers are considered for screening AD, detection of some includes invasive methods, while others are based on neuroimaging and neuropsychological cognitive tasks, such as the Montreal Cognitive Assessment. While imaging and brain signal acquisition modalities demand centralized facilities, making them expensive and time-consuming, invasive sample collection procedures are prohibitory.
While the current non-invasive brain wave monitoring techniques, such as electroencephalograms (EEG) or brain imaging modalities, contribute significantly in depicting manifestations of disease, there still exists no gold standard clinical non-invasive solution for disease progression tracking. To this end, different eye segments can be evaluated for discovering AD-associated representations, including the retina, iris, or simply eye movement. Pathologies of AD in the neurosensory layers of the retina are evident, and hence, interest in the extraction of retinal Optical Coherence Tomography (OCT) features representative of the disease progression and in the early stages of cognitive impairment is growing. The biological components in retina tissue, compared to non-impaired individuals, provide further support for the utility of the retina in AD diagnosis. Consequently using OCT images for this purpose would be of paramount value, even in mild cases, upon discovery of the AD-specific features in the retina and optic nerve. The efforts to find such patterns distinguishing AD patients and the age-matched cohort, have been carried out since 2001, with instances such as retinal nerve fiber layer (rNFL) thinning in AD patients revealed. As another non-invasive approach, eye tracking is also of interest, since it serves as a language and culturally independent method for large-scale screening, and has been previously shown to be effective for distinguishing patients with neurological disorders like autism spectrum disorder (ASD). In this project, we first will conduct a clinical study to evaluate the clinical value and efficacy of eye-movement tracking, OCT features, and EEG signals in discovering patterns distinguishing AD patients from the control group. The data from this study will assist in determining the most discriminative AD markers in a first-time biomarker discovery initiative establishing correlations between markers, in particular with more established OCT features. In the second phase, we would undertake a “panel” approach, i.e. simultaneous detection of serval markers. “Smart goggles” will be developed with embedded cameras for eye tracking with extended electrodes on the temporal lobes for scalp EEG acquisition. The precise specifications of the final location of electrodes, their types, and hence the characteristics of the signals will be determined based on the outcomes of the first phase of the study. The data acquired from all these sensory modalities will be wirelessly communicated to a portable processor (e.g., a cellphone), and analyzed via Machine Learning (ML) algorithms for automated assessment of AD progression. This integrated wearable system will serve as a Point-of-Care (PoC) platform, enabling early detection of AD, and monitoring the effect of treatments on AD prognosis.
Duties and Responsibilities: Assist in acquiring OCT images, eye movement tracking, and brain signals from the recruited AD patients and age-matched healthy individuals. Assist in developing machine learning (ML) algorithms for feature extraction in the collected data and providing correlations and specific markers associated with AD. Assist in the design and development of smart goggles for recording image-based and temporal sensory information. Assist in functionality assessment and optimization of the developed smart goggles and a pilot study in a clinical cohort for evaluating the utility of the system in detecting AD cases.
Desired Technical Skills: Background in Electrical, Computer, and Software Engineering: Demonstrated expertise in the fundamentals and advanced concepts of these engineering disciplines, including systems design, programming, and hardware-software integration. Experience in AI Model Development: Skilled in training machine learning models, labeling datasets, and programming for AI applications. This includes hands-on experience in developing, supervising, and evaluating the performance of AI models across various applications. Experience in Circuit Design: Knowledge in designing circuits specifically for EEG electrodes, including aspects of miniaturization, signal integrity, and noise reduction. Programming Skills: Experience in programming languages commonly used in Electrical, Computer, and Software Engineering, such as Python, C++, Java, and MATLAB. Team Collaboration: Experience in working collaboratively in interdisciplinary teams that involve complex electrical, computer, and software engineering tasks.
Desired Course(s): N/A.
Other Desired Qualifications: N/A.
Professor: Marios Fokaefs
Contact Info: fokaefs@yorku.ca
Lab Website: https://fokaefs.github.io/
Position Type: Lassonde Undergraduate Research Award (LURA); NSERC Undergraduate Student Research Award (USRA)
Open Positions: 3
Project Description: Generative AI has attracted a lot of attention recently from the research community. Its ability to generate complex solutions from similar examples has made it an interesting solution for problems that require a certain degree of creativity. Design processes that are necessary in software architecture require similarly creative solutions. In addition, the problem of translating natural text requirements into design and architecture artifacts is non-linear and in most cases unstructured. In this sense, design processes can also benefit from the use of Large Language Models (LLM), known for their ability to understand and generate natural language. The objective of this project is to explore the ability of LLMs and generative AI to produce functional and performant software architectures given free-text requirement descriptions. The developed models will need to parse these descriptions, extract functional and non-functional requirements, and produce architectural and high-level designs, as UML diagrams, following proper design principles and patterns, and architectural styles. Explainability and justifiability are of utmost importance for the produced designs.
Duties and Responsibilities: Become familiar with tools and technologies such as ChatGPT and deep learning. Design and optimize queries on software architectures. Evaluate the results of ChatGPT with respect to architectural patterns and quality. Train open-source Generative AI tools specifically for software architecture problems.
Desired Technical Skills: Good programming skills, preferably in Java or Python. Good knowledge on software architecture, software design, distributed and cloud systems. Adequate knowledge on AI, machine learning, deep learning, and relevant technologies.
Desired Course(s): LE/EECS 3311 3.00 – Software Design, LE/EECS 3342 3.00 – System Specification and Refinement, LE/EECS 3401 3.00 – Introduction to Artificial Intelligence and Logic Programming, LE/EECS 4404 3.00 – Introduction to Machine Learning and Pattern Recognition, LE/EECS 4412 3.00 – Data Mining
Other Desired Qualifications: N/A.
Contact Info: fokaefs@yorku.ca
Lab Website: https://fokaefs.github.io/
Position Type: Lassonde Undergraduate Research Award (LURA); NSERC Undergraduate Student Research Award (USRA)
Open Positions: 3
Project Description: Generative AI has attracted a lot of attention recently from the research community. Its ability to generate complex solutions from similar examples has made it an interesting solution for problems that require a certain degree of creativity. Design processes that are necessary in software architecture require similarly creative solutions. In addition, the problem of translating natural text requirements into design and architecture artifacts is non-linear and in most cases unstructured. In this sense, design processes can also benefit from the use of Large Language Models (LLM), known for their ability to understand and generate natural language. The objective of this project is to explore the ability of LLMs and generative AI to produce functional and performant software architectures given free-text requirement descriptions. The developed models will need to parse these descriptions, extract functional and non-functional requirements, and produce architectural and high-level designs, as UML diagrams, following proper design principles and patterns, and architectural styles. Explainability and justifiability are of utmost importance for the produced designs.
Duties and Responsibilities: Become familiar with tools and technologies such as ChatGPT and deep learning. Design and optimize queries on software architectures. Evaluate the results of ChatGPT with respect to architectural patterns and quality. Train open-source Generative AI tools specifically for software architecture problems.
Desired Technical Skills: Good programming skills, preferably in Java or Python. Good knowledge on software architecture, software design, distributed and cloud systems. Adequate knowledge on AI, machine learning, deep learning, and relevant technologies.
Desired Course(s): LE/EECS 3311 3.00 – Software Design, LE/EECS 3342 3.00 – System Specification and Refinement, LE/EECS 3401 3.00 – Introduction to Artificial Intelligence and Logic Programming, LE/EECS 4404 3.00 – Introduction to Machine Learning and Pattern Recognition, LE/EECS 4412 3.00 – Data Mining
Other Desired Qualifications: N/A.
Professor: Marios Fokaefs
Contact Info: fokaefs@yorku.ca
Lab Website: https://fokaefs.github.io/
Position Type: Lassonde Undergraduate Research Award (LURA); NSERC Undergraduate Student Research Award (USRA)
Open Positions: 1
Project Description: Machine Learning and Artificial Intelligence in general are slowly becoming an integral part in the majority of software systems and applications. In combination with the advancements in pervasive and ubiquitous computing, it is now implied that AI models need to be deployable on any device. However, it is not clear yet what the implications are of low capacity devices, such as mobile phones, on the deployment, performance, and accuracy of AI models. In this project, we aim to firstly explore these implications and quantify them, in order to guide the optimization of AI models in terms of resource consumption and constrained by requirements on their performance and accuracy. Next, we will integrate these considerations in MLOps pipelines to automate the optimization and make it an integral part of the model’s lifecycle. The expected outcome of the project is the development of models to simultaneously capture the aspect of the AI models as software artifacts (performance and resource consumption) and as AI artifacts (accuracy). Based on these models, we will also develop tools for the optimal and adaptive deployment of AI models on low capacity devices.
Duties and Responsibilities: In the context of this project, the student will work with an existing prototype of an Android mobile application that receives and analyzes personalized tests (for COVID-19, glucose, pregnancy, and others). The objective is to redesign the machine learning model that analyzes the test so that it can be deployed and used locally in the mobile device. In parallel, the student will continue developing the app with new features to support users as well as user studies for research.
Desired Technical Skills: Good software development skills, preferably in Java, Kotlin and/or Python. Good knowledge about machine learning and relevant tools like TensorFlow. Adequate understanding of software lifecycle models, such as DevOps, MLOps etc.
Desired Course(s): LE/EECS 3311 3.00 – Software Design, LE/EECS 4404 3.00 – Introduction to Machine Learning and Pattern Recognition, LE/EECS 4412 3.00 – Data Mining, LE/EECS 4443 3.00 – Mobile User interfaces (the courses are recommendations and they will be considered only as an advantage, but not as a strict requirement)
Other Desired Qualifications: N/A.
Contact Info: fokaefs@yorku.ca
Lab Website: https://fokaefs.github.io/
Position Type: Lassonde Undergraduate Research Award (LURA); NSERC Undergraduate Student Research Award (USRA)
Open Positions: 1
Project Description: Machine Learning and Artificial Intelligence in general are slowly becoming an integral part in the majority of software systems and applications. In combination with the advancements in pervasive and ubiquitous computing, it is now implied that AI models need to be deployable on any device. However, it is not clear yet what the implications are of low capacity devices, such as mobile phones, on the deployment, performance, and accuracy of AI models. In this project, we aim to firstly explore these implications and quantify them, in order to guide the optimization of AI models in terms of resource consumption and constrained by requirements on their performance and accuracy. Next, we will integrate these considerations in MLOps pipelines to automate the optimization and make it an integral part of the model’s lifecycle. The expected outcome of the project is the development of models to simultaneously capture the aspect of the AI models as software artifacts (performance and resource consumption) and as AI artifacts (accuracy). Based on these models, we will also develop tools for the optimal and adaptive deployment of AI models on low capacity devices.
Duties and Responsibilities: In the context of this project, the student will work with an existing prototype of an Android mobile application that receives and analyzes personalized tests (for COVID-19, glucose, pregnancy, and others). The objective is to redesign the machine learning model that analyzes the test so that it can be deployed and used locally in the mobile device. In parallel, the student will continue developing the app with new features to support users as well as user studies for research.
Desired Technical Skills: Good software development skills, preferably in Java, Kotlin and/or Python. Good knowledge about machine learning and relevant tools like TensorFlow. Adequate understanding of software lifecycle models, such as DevOps, MLOps etc.
Desired Course(s): LE/EECS 3311 3.00 – Software Design, LE/EECS 4404 3.00 – Introduction to Machine Learning and Pattern Recognition, LE/EECS 4412 3.00 – Data Mining, LE/EECS 4443 3.00 – Mobile User interfaces (the courses are recommendations and they will be considered only as an advantage, but not as a strict requirement)
Other Desired Qualifications: N/A.
Professor: Hina Tabassum
Contact Info: hinat@yorku.ca
Lab Website: https://lassonde.yorku.ca/users/hina
Position Type: Lassonde Undergraduate Research Award (LURA); NSERC Undergraduate Student Research Award (USRA)
Open Positions: 1
Project Description: Vehicle-to-Infrastructure (V2I) communication is becoming critical for the enhanced reliability of Autonomous Vehicles (AVs). However, the uncertainties in the road-traffic and AVs’ wireless connections can severely impair timely decision-making. It is thus critical to simultaneously optimize the AVs’ network selection and driving policies in order to minimize road collisions while maximizing the reliability and latency of communication links. The impact of speed on the communication data rates as well as road traffic flow will be explored considering intelligent driver models. The problem will be formulated as a multi-objective Markov Decision Process (MDP); then the performance of (1) meta-reinforcement learning, and (2) quantum-inspired deep reinforcement learning solutions will be explored. The developed solutions will be compared to several representative reinforcement learning solutions. The proposed framework is designed to maximize the traffic flow and minimize collisions by controlling the vehicle’s motion dynamics (i.e., speed and acceleration) from autonomous driving perspective, and maximize the data rates and minimize handoffs by jointly controlling the vehicle’s motion dynamics and network selection from telecommunication perspective.
Duties and Responsibilities: Explore the feasibility and performance of quantum-inspired machine learning methods in wireless network resource management.
Desired Technical Skills: Programming skills (Python, MATLAB) – Familiarity with wireless communications, digital communications, and deep learning concepts preferred.
Desired Course(s): Courses related to Basics of Machine learning, Optimization theory, and Wireless communications would be preferred.
Other Desired Qualifications: N/A.
Contact Info: hinat@yorku.ca
Lab Website: https://lassonde.yorku.ca/users/hina
Position Type: Lassonde Undergraduate Research Award (LURA); NSERC Undergraduate Student Research Award (USRA)
Open Positions: 1
Project Description: Vehicle-to-Infrastructure (V2I) communication is becoming critical for the enhanced reliability of Autonomous Vehicles (AVs). However, the uncertainties in the road-traffic and AVs’ wireless connections can severely impair timely decision-making. It is thus critical to simultaneously optimize the AVs’ network selection and driving policies in order to minimize road collisions while maximizing the reliability and latency of communication links. The impact of speed on the communication data rates as well as road traffic flow will be explored considering intelligent driver models. The problem will be formulated as a multi-objective Markov Decision Process (MDP); then the performance of (1) meta-reinforcement learning, and (2) quantum-inspired deep reinforcement learning solutions will be explored. The developed solutions will be compared to several representative reinforcement learning solutions. The proposed framework is designed to maximize the traffic flow and minimize collisions by controlling the vehicle’s motion dynamics (i.e., speed and acceleration) from autonomous driving perspective, and maximize the data rates and minimize handoffs by jointly controlling the vehicle’s motion dynamics and network selection from telecommunication perspective.
Duties and Responsibilities: Explore the feasibility and performance of quantum-inspired machine learning methods in wireless network resource management.
Desired Technical Skills: Programming skills (Python, MATLAB) – Familiarity with wireless communications, digital communications, and deep learning concepts preferred.
Desired Course(s): Courses related to Basics of Machine learning, Optimization theory, and Wireless communications would be preferred.
Other Desired Qualifications: N/A.
Professor: Eric Ruppert
Contact Info: ruppert@eecs.yorku.ca
Lab Website: www.eecs.yorku.ca/~eruppert
Position Type: Lassonde Undergraduate Research Award (LURA); NSERC Undergraduate Student Research Award (USRA)
Open Positions: 1
Project Description: This project is concerned with the development of concurrent data structures, which can be accessed simultaneously by multiple processes. These data structures are essential for harnessing the full power of modern multicore architectures. Processes must carefully coordinate their accesses to the data structures to ensure that simultaneous updates do not cause shared data to become corrupted. The goal of the project is to develop new concurrent data structures with good time and space complexity.
Duties and Responsibilities: Some background reading will be required, followed by work on the development of a new data structure and a proof of correctness for it.
Desired Technical Skills: Ability to prove correctness of algorithms and analyse their complexity.
Desired Course(s): Computer science or software engineering students who have excelled in courses in theoretical computer science (e.g., data structures, algorithms, automata theory, complexity).
Other Desired Qualifications: Strong math background.
Contact Info: ruppert@eecs.yorku.ca
Lab Website: www.eecs.yorku.ca/~eruppert
Position Type: Lassonde Undergraduate Research Award (LURA); NSERC Undergraduate Student Research Award (USRA)
Open Positions: 1
Project Description: This project is concerned with the development of concurrent data structures, which can be accessed simultaneously by multiple processes. These data structures are essential for harnessing the full power of modern multicore architectures. Processes must carefully coordinate their accesses to the data structures to ensure that simultaneous updates do not cause shared data to become corrupted. The goal of the project is to develop new concurrent data structures with good time and space complexity.
Duties and Responsibilities: Some background reading will be required, followed by work on the development of a new data structure and a proof of correctness for it.
Desired Technical Skills: Ability to prove correctness of algorithms and analyse their complexity.
Desired Course(s): Computer science or software engineering students who have excelled in courses in theoretical computer science (e.g., data structures, algorithms, automata theory, complexity).
Other Desired Qualifications: Strong math background.
Professor: Amirali Amirsoleimani
Contact Info: amirsol@yorku.ca
Lab Website: https://lcrain.eecs.yorku.ca/
Position Type: NSERC Undergraduate Student Research Award (USRA)
Open Positions: 1
Project Description: We are looking to design a new brain-inspired digital platform by using neuromorphic computing science. The proposed system is intended to be a part of a larger system for biological signal processing applications. We are going to use the state-of the art techniques for training the system and inference using a high-end FPGA boards (Xilinx family). The final results are planed to be published as a scientific journal or conference paper.
Duties and Responsibilities: 1. Developing the model, design, and analysis of the neuromorphic cyber physical system. 2. Implementation and synthesis on the FPGA board. 3. Verification and testing the system for specific application. 4. Training and inference will be done on the FPGA board. 5. Results needs to be placed in a body of a scientific paper.
Desired Technical Skills: Has knowledge and experience on Verilog and Digital Design (having previous design implementations will be an asset), has a good understanding of machine learning and neural networks, and is interested in learning new scientific domains like neuromorphic computing. Being a fast learner is also an asset.
Desired Course(s): Computer Engineering, Computer Science, Electrical Engineering, Machine Learning and Artificial Intelligence, Digital logic design, Embedded systems, Cyber Physical Systems.
Other Desired Qualifications: The student should be able to communicate with other lab members and can work as a team member in the research group.
Contact Info: amirsol@yorku.ca
Lab Website: https://lcrain.eecs.yorku.ca/
Position Type: NSERC Undergraduate Student Research Award (USRA)
Open Positions: 1
Project Description: We are looking to design a new brain-inspired digital platform by using neuromorphic computing science. The proposed system is intended to be a part of a larger system for biological signal processing applications. We are going to use the state-of the art techniques for training the system and inference using a high-end FPGA boards (Xilinx family). The final results are planed to be published as a scientific journal or conference paper.
Duties and Responsibilities: 1. Developing the model, design, and analysis of the neuromorphic cyber physical system. 2. Implementation and synthesis on the FPGA board. 3. Verification and testing the system for specific application. 4. Training and inference will be done on the FPGA board. 5. Results needs to be placed in a body of a scientific paper.
Desired Technical Skills: Has knowledge and experience on Verilog and Digital Design (having previous design implementations will be an asset), has a good understanding of machine learning and neural networks, and is interested in learning new scientific domains like neuromorphic computing. Being a fast learner is also an asset.
Desired Course(s): Computer Engineering, Computer Science, Electrical Engineering, Machine Learning and Artificial Intelligence, Digital logic design, Embedded systems, Cyber Physical Systems.
Other Desired Qualifications: The student should be able to communicate with other lab members and can work as a team member in the research group.
Professor: Meiying Qin
Contact Info: mqin@yorku.ca
Lab Website: https://lassonde.yorku.ca/users/mqin
Position Type: NSERC Undergraduate Student Research Award (USRA)
Open Positions: 2
Project Description: The project aims to help first-year students transition from high school to university and understand university expectations by developing a simulation game. This game provides incoming students with a comprehensive preview of life as a computer science student, allowing them to experience university life before classes begin. The goal is to replace traditional advice with experiential learning, helping students internalize important insights. Understanding these expectations is intended to facilitate a smoother transition, contributing to a less stressful first year and supporting overall well-being.
In the project, you will have the opportunity to contribute to the creative design of the game, refining your soft engineering skills and gaining experience in game development.
Duties and Responsibilities: Game Design; Implementing the Game; Game Testing.
Desired Technical Skills: If you are a CS student, you should be strong in programming and you are not required to have game development experience; If you are a student in digital media, you should have strong skills in game development.
Desired Course(s): Students in the EECS program and digital media are all welcome to apply.
If you are a CS student, you should complete a course in object-oriented programming. Having experience in the industry as a software engineer is preferred but not required (e.g., internship).
If you are a digital media student, you should complete the courses in game design I and II. Having completed the game mechanics course and/or a project course is preferred but not required.
Other Desired Qualifications: N/A.
Contact Info: mqin@yorku.ca
Lab Website: https://lassonde.yorku.ca/users/mqin
Position Type: NSERC Undergraduate Student Research Award (USRA)
Open Positions: 2
Project Description: The project aims to help first-year students transition from high school to university and understand university expectations by developing a simulation game. This game provides incoming students with a comprehensive preview of life as a computer science student, allowing them to experience university life before classes begin. The goal is to replace traditional advice with experiential learning, helping students internalize important insights. Understanding these expectations is intended to facilitate a smoother transition, contributing to a less stressful first year and supporting overall well-being.
In the project, you will have the opportunity to contribute to the creative design of the game, refining your soft engineering skills and gaining experience in game development.
Duties and Responsibilities: Game Design; Implementing the Game; Game Testing.
Desired Technical Skills: If you are a CS student, you should be strong in programming and you are not required to have game development experience; If you are a student in digital media, you should have strong skills in game development.
Desired Course(s): Students in the EECS program and digital media are all welcome to apply.
If you are a CS student, you should complete a course in object-oriented programming. Having experience in the industry as a software engineer is preferred but not required (e.g., internship).
If you are a digital media student, you should complete the courses in game design I and II. Having completed the game mechanics course and/or a project course is preferred but not required.
Other Desired Qualifications: N/A.
Professor: Meiying Qin
Contact Info: mqin@yorku.ca
Lab Website: https://lassonde.yorku.ca/users/mqin
Position Type: NSERC Undergraduate Student Research Award (USRA)
Open Positions: 2
Project Description: Debugging is one of the most important skills for computer science students. However, first-year students are usually not comfortable with working with a debugger. In order to help ease the process for first-year students, we plan to write an application that can visualize the process by animating the variable manipulated using mixed/augmented reality.
In this project, you will have the opportunity to gain hands-on experience in both designing and implementing a software application. You will gain experience in mixed/augmented reality.
Duties and Responsibilities: Design the application; Implement the application; Test the application.
Desired Technical Skills: Strong programming skills or experience in developing MR/AR applications. While possessing both skills is not mandatory, it is considered advantageous.
Desired Course(s): At least a course in object-oriented programming.
Other Desired Qualifications: N/A.
Contact Info: mqin@yorku.ca
Lab Website: https://lassonde.yorku.ca/users/mqin
Position Type: NSERC Undergraduate Student Research Award (USRA)
Open Positions: 2
Project Description: Debugging is one of the most important skills for computer science students. However, first-year students are usually not comfortable with working with a debugger. In order to help ease the process for first-year students, we plan to write an application that can visualize the process by animating the variable manipulated using mixed/augmented reality.
In this project, you will have the opportunity to gain hands-on experience in both designing and implementing a software application. You will gain experience in mixed/augmented reality.
Duties and Responsibilities: Design the application; Implement the application; Test the application.
Desired Technical Skills: Strong programming skills or experience in developing MR/AR applications. While possessing both skills is not mandatory, it is considered advantageous.
Desired Course(s): At least a course in object-oriented programming.
Other Desired Qualifications: N/A.
Professor: Franck van Breugel and Eric Ruppert
Contact Info: franck@yorku.ca
Lab Website: www.eecs.yorku.ca/~franck
Position Type: Lassonde Undergraduate Research Award (LURA); NSERC Undergraduate Student Research Award (USRA)
Open Positions: 2
Project Description: Randomness plays a key role in numerous algorithms in various fields, ranging from machine learning to cryptography. Finding bugs in implementations of algorithms is normally done by testing the code. However, with randomness involved, testing is less effective. For example, consider the following Java code.
import java.util.Random;
public class Example {
public static void main(String[] args) {
Random random = new Random();
System.out.print(1 / random.nextInt(9));
}
}
The above application may result in ten different executions, since it randomly chooses an integer in the interval [0, 9]. In 80% of the cases, the application prints zero, in 10% it prints one, and in the remaining 10% it crashes because of an uncaught exception due to a division by zero. Of course, it may take more than ten executions before we encounter the exception. In case we choose an integer in the interval [0, 999999] it may take many executions before encountering the exception. For example, if we execute the application one million times, there is still a 37% chance that we do not encounter the exception.
To find bugs in implementations of randomized algorithms, probabilistic model checking is often used. First, a model of the code is built. For example, NASA’s Java PathFinder builds a model from Java code and our extension jpf-probabilistic keeps track of the probabilities associated with random choices such as random.nextInt(9). Next, properties, such as there are no uncaught exceptions, of the model are checked. For example, PRISM, a tool developed at the University of Oxford, checks such properties. However, if the model is large then checking such properties may take too much time or memory. To address this, the model is minimized by identifying states that give rise to the same behaviour resulting often in a much smaller model before checking the property. Research has shown that the time it takes to minimize the model and subsequently check the property in the minimized model is often significantly smaller than the time it takes to check the property in the original model.
Duties and Responsibilities: This summer research project extends the above described research in several directions. In particular, we will consider; 1) many more models than the original research, 2) several tools to check properties: PRISM, MRMC, and Storm, and 3) several algorithms to minimize the model, including a brand new, unpublished algorithm.
Desired Technical Skills: To contribute to this project, you should; 1) find the project fascinating, 2) be comfortable in Java, 3) be familiar with various algorithms, and 4) enjoy working in a group.
Desired Course(s): A course on Java and a course on algorithms and/or data structures.
Other Desired Qualifications: N/A.
Contact Info: franck@yorku.ca
Lab Website: www.eecs.yorku.ca/~franck
Position Type: Lassonde Undergraduate Research Award (LURA); NSERC Undergraduate Student Research Award (USRA)
Open Positions: 2
Project Description: Randomness plays a key role in numerous algorithms in various fields, ranging from machine learning to cryptography. Finding bugs in implementations of algorithms is normally done by testing the code. However, with randomness involved, testing is less effective. For example, consider the following Java code.
import java.util.Random;
public class Example {
public static void main(String[] args) {
Random random = new Random();
System.out.print(1 / random.nextInt(9));
}
}
The above application may result in ten different executions, since it randomly chooses an integer in the interval [0, 9]. In 80% of the cases, the application prints zero, in 10% it prints one, and in the remaining 10% it crashes because of an uncaught exception due to a division by zero. Of course, it may take more than ten executions before we encounter the exception. In case we choose an integer in the interval [0, 999999] it may take many executions before encountering the exception. For example, if we execute the application one million times, there is still a 37% chance that we do not encounter the exception.
To find bugs in implementations of randomized algorithms, probabilistic model checking is often used. First, a model of the code is built. For example, NASA’s Java PathFinder builds a model from Java code and our extension jpf-probabilistic keeps track of the probabilities associated with random choices such as random.nextInt(9). Next, properties, such as there are no uncaught exceptions, of the model are checked. For example, PRISM, a tool developed at the University of Oxford, checks such properties. However, if the model is large then checking such properties may take too much time or memory. To address this, the model is minimized by identifying states that give rise to the same behaviour resulting often in a much smaller model before checking the property. Research has shown that the time it takes to minimize the model and subsequently check the property in the minimized model is often significantly smaller than the time it takes to check the property in the original model.
Duties and Responsibilities: This summer research project extends the above described research in several directions. In particular, we will consider; 1) many more models than the original research, 2) several tools to check properties: PRISM, MRMC, and Storm, and 3) several algorithms to minimize the model, including a brand new, unpublished algorithm.
Desired Technical Skills: To contribute to this project, you should; 1) find the project fascinating, 2) be comfortable in Java, 3) be familiar with various algorithms, and 4) enjoy working in a group.
Desired Course(s): A course on Java and a course on algorithms and/or data structures.
Other Desired Qualifications: N/A.
Professor: Meiying Qin
Contact Info: mqin@yorku.ca
Lab Website: https://lassonde.yorku.ca/users/mqin
Position Type: NSERC Undergraduate Student Research Award (USRA)
Open Positions: 2
Project Description: Robot tutors have demonstrated efficacy in aiding students, yet their widespread adoption faces hurdles due to high costs and limited scalability. Current robot tutors are often impractical for widespread use in universities or personal adoption due to their expense. This project seeks to overcome these limitations by developing an affordable physical robot tutor. The objective is to create a solution that meets the educational needs of students without imposing financial constraints. By focusing on cost-effectiveness, this initiative aims to make advanced tutoring technology accessible on a broader scale, fostering inclusive and impactful learning experiences.
Duties and Responsibilities: Build a physical robot.
Desired Technical Skills: Experience working with hardware such as motors, sensors, and mechatronics. Knowledge of ROS (e.g., robot operating system), and 3D printing/CAD modeling, is recommended but not required.
Desired Course(s): Introduction to robotics or similar courses. If you have experience building robots, the course requirement can be waived.
Other Desired Qualifications: N/A.
Contact Info: mqin@yorku.ca
Lab Website: https://lassonde.yorku.ca/users/mqin
Position Type: NSERC Undergraduate Student Research Award (USRA)
Open Positions: 2
Project Description: Robot tutors have demonstrated efficacy in aiding students, yet their widespread adoption faces hurdles due to high costs and limited scalability. Current robot tutors are often impractical for widespread use in universities or personal adoption due to their expense. This project seeks to overcome these limitations by developing an affordable physical robot tutor. The objective is to create a solution that meets the educational needs of students without imposing financial constraints. By focusing on cost-effectiveness, this initiative aims to make advanced tutoring technology accessible on a broader scale, fostering inclusive and impactful learning experiences.
Duties and Responsibilities: Build a physical robot.
Desired Technical Skills: Experience working with hardware such as motors, sensors, and mechatronics. Knowledge of ROS (e.g., robot operating system), and 3D printing/CAD modeling, is recommended but not required.
Desired Course(s): Introduction to robotics or similar courses. If you have experience building robots, the course requirement can be waived.
Other Desired Qualifications: N/A.
Professor: Yves Lesperance
Contact Info: lesperan@yorku.ca
Lab Website: https://www.eecs.yorku.ca/~lesperan/
Position Type: Lassonde Undergraduate Research Award (LURA); NSERC Undergraduate Student Research Award (USRA)
Open Positions: 1
Project Description: Reasoning about action (RAC), including generating plans to achieve goals, is a key capability for autonomous agents. Mainstream techniques for RAC and automated planning (based on heuristic searches), are very effective, but they rely on a human modeler specifying the dynamic domain and queries/goals formally. Some recent research has investigated whether pretrained Language Models (LM) can effectively reason about action and change while avoiding the need to formally specify the domain. For instance, He et al. (ACL 2023) has studied the performance of some LMs on fundamental RAC tasks such as Projection, Executability, Plan Verification, and Goal Recognition. The LMs were first fine-tuned/pretrained on Blocks World domains and task instances (with a STRIPS semantics) and then tested on new instances. The LMs performed rather well on similar instances, but generalized poorly to tasks involving longer action sequences or more domain objects. In this project, the student will use the datasets generated in this work to experiment with newer language models and various fine-tuning methods to see if generalization can be improved.
Duties and Responsibilities: Complete some background reading. Select LM(s) (e.g. Llama, Mistral) and fine-tuning method(s) (e.g., PEFT) to be used. Use the data form He et al. to train/fine-tune the LM. Then evaluate generalization performance on He et al’s generalization dataset. Write report. If there is time, generate data from a new planning domain and/or investigate LMs capabilities for hierarchical planning, where the planner uses knowledge about how to decompose the goal/task into subgoals/tasks.
Desired Technical Skills: Has Python programming skills. Some previous exposure to machine learning, automated planning, and first-order logic, is desirable.
Desired Course(s): Enrolled in CS/CE/SE program. Has taken LE/EECS 3401 3.00 – Introduction to Artificial Intelligence and Logic Programming.
Other Desired Qualifications: Familiarity with Large Language Models or Natural Language Processing and relevant technologies / tools is an asset.
Contact Info: lesperan@yorku.ca
Lab Website: https://www.eecs.yorku.ca/~lesperan/
Position Type: Lassonde Undergraduate Research Award (LURA); NSERC Undergraduate Student Research Award (USRA)
Open Positions: 1
Project Description: Reasoning about action (RAC), including generating plans to achieve goals, is a key capability for autonomous agents. Mainstream techniques for RAC and automated planning (based on heuristic searches), are very effective, but they rely on a human modeler specifying the dynamic domain and queries/goals formally. Some recent research has investigated whether pretrained Language Models (LM) can effectively reason about action and change while avoiding the need to formally specify the domain. For instance, He et al. (ACL 2023) has studied the performance of some LMs on fundamental RAC tasks such as Projection, Executability, Plan Verification, and Goal Recognition. The LMs were first fine-tuned/pretrained on Blocks World domains and task instances (with a STRIPS semantics) and then tested on new instances. The LMs performed rather well on similar instances, but generalized poorly to tasks involving longer action sequences or more domain objects. In this project, the student will use the datasets generated in this work to experiment with newer language models and various fine-tuning methods to see if generalization can be improved.
Duties and Responsibilities: Complete some background reading. Select LM(s) (e.g. Llama, Mistral) and fine-tuning method(s) (e.g., PEFT) to be used. Use the data form He et al. to train/fine-tune the LM. Then evaluate generalization performance on He et al’s generalization dataset. Write report. If there is time, generate data from a new planning domain and/or investigate LMs capabilities for hierarchical planning, where the planner uses knowledge about how to decompose the goal/task into subgoals/tasks.
Desired Technical Skills: Has Python programming skills. Some previous exposure to machine learning, automated planning, and first-order logic, is desirable.
Desired Course(s): Enrolled in CS/CE/SE program. Has taken LE/EECS 3401 3.00 – Introduction to Artificial Intelligence and Logic Programming.
Other Desired Qualifications: Familiarity with Large Language Models or Natural Language Processing and relevant technologies / tools is an asset.
Professor: Andrew Eckford
Contact Info: aeckford@yorku.ca
Lab Website: http://eckfordlab.org/
Position Type: Lassonde Undergraduate Research Award (LURA); NSERC Undergraduate Student Research Award (USRA)
Open Positions: 1
Project Description: This research project aims to investigate the applicability of Kelly betting strategy in elucidating logistic growth dynamics, transcending traditional boundaries by implementing a simulation titled the “Game of Fitness.” Drawing inspiration from Conway’s Game of Life, this project endeavors to model the growth of biological organisms, yet its broader implications extend to diverse fields, including finance. The primary objective is to empirically test the hypothesis that logistic growth phenomena can be effectively explained through the lens of a Kelly betting strategy. The proposed simulation will serve as a dynamic platform to explore the intricate relationship between betting strategies and the emergence of logistic growth patterns. By delving into this interdisciplinary realm, our research aims to provide insights into the universal principles underlying growth processes and, consequently, contribute to the advancement of both biological and financial sciences. The key focus of this project lies in the implementation of the Game of Fitness simulation, marking a significant step toward unraveling the complexities of logistic growth through a novel and versatile computational approach.
Duties and Responsibilities: The student will be responsible for writing and running simulation code that implements the project objectives. The student will work closely with the PI and an international researcher to implement the project details.
Desired Technical Skills: Strong programming skills, especially in Python, are a requirement. Familiarity with Jupyter Notebook is a plus.
Desired Course(s): A background in probability and statistics is a plus.
Other Desired Qualifications: N/A.
Contact Info: aeckford@yorku.ca
Lab Website: http://eckfordlab.org/
Position Type: Lassonde Undergraduate Research Award (LURA); NSERC Undergraduate Student Research Award (USRA)
Open Positions: 1
Project Description: This research project aims to investigate the applicability of Kelly betting strategy in elucidating logistic growth dynamics, transcending traditional boundaries by implementing a simulation titled the “Game of Fitness.” Drawing inspiration from Conway’s Game of Life, this project endeavors to model the growth of biological organisms, yet its broader implications extend to diverse fields, including finance. The primary objective is to empirically test the hypothesis that logistic growth phenomena can be effectively explained through the lens of a Kelly betting strategy. The proposed simulation will serve as a dynamic platform to explore the intricate relationship between betting strategies and the emergence of logistic growth patterns. By delving into this interdisciplinary realm, our research aims to provide insights into the universal principles underlying growth processes and, consequently, contribute to the advancement of both biological and financial sciences. The key focus of this project lies in the implementation of the Game of Fitness simulation, marking a significant step toward unraveling the complexities of logistic growth through a novel and versatile computational approach.
Duties and Responsibilities: The student will be responsible for writing and running simulation code that implements the project objectives. The student will work closely with the PI and an international researcher to implement the project details.
Desired Technical Skills: Strong programming skills, especially in Python, are a requirement. Familiarity with Jupyter Notebook is a plus.
Desired Course(s): A background in probability and statistics is a plus.
Other Desired Qualifications: N/A.
Professor: Hina Tabassum
Contact Info: hinat@yorku.ca
Lab Website: https://lassonde.yorku.ca/users/hina
Position Type: Lassonde Undergraduate Research Award (LURA); NSERC Undergraduate Student Research Award (USRA)
Open Positions: 1
Project Description: Generative Adversarial Networks (GANs) represent a category of Machine Learning (ML) algorithms capable of addressing challenges related to wireless channel prediction and competitive spectrum, power, antenna allocation to minimize energy consumption related to the . GANs offer several advantages in this context. Firstly, they can learn and synthesize field data, saving both time and resources compared to traditional methods. Secondly, they support the pre-training of classifiers through the utilization of semi-supervised data. Thirdly, they contribute to increased resolution. This project will investigate various GAN types and performance metrics for GANs in wireless communications and wireless sensing applications.
Duties and Responsibilities: The student will review state-of-the-art and relevant GAN models that can be applied for wireless applications. The student will then comparatively analyze the performance of various GAN models.
Desired Technical Skills: The students should possess algorithmic design and development knowledge, as well as demonstrated strong programming skills.
Desired Course(s): It is recommended to have completed some of the communications courses such as LE/EECS 4214 4.00 – Digital Communications, LE/EECS 3213 3.00 – Communication Networks, LE/EECS 4215 3.00 – Mobile Communications, LE/EECS 4404 3.00 – Introduction to Machine Learning and Pattern Recognition, etc.
Other Desired Qualifications: Other qualifications include good communication skills.
Contact Info: hinat@yorku.ca
Lab Website: https://lassonde.yorku.ca/users/hina
Position Type: Lassonde Undergraduate Research Award (LURA); NSERC Undergraduate Student Research Award (USRA)
Open Positions: 1
Project Description: Generative Adversarial Networks (GANs) represent a category of Machine Learning (ML) algorithms capable of addressing challenges related to wireless channel prediction and competitive spectrum, power, antenna allocation to minimize energy consumption related to the . GANs offer several advantages in this context. Firstly, they can learn and synthesize field data, saving both time and resources compared to traditional methods. Secondly, they support the pre-training of classifiers through the utilization of semi-supervised data. Thirdly, they contribute to increased resolution. This project will investigate various GAN types and performance metrics for GANs in wireless communications and wireless sensing applications.
Duties and Responsibilities: The student will review state-of-the-art and relevant GAN models that can be applied for wireless applications. The student will then comparatively analyze the performance of various GAN models.
Desired Technical Skills: The students should possess algorithmic design and development knowledge, as well as demonstrated strong programming skills.
Desired Course(s): It is recommended to have completed some of the communications courses such as LE/EECS 4214 4.00 – Digital Communications, LE/EECS 3213 3.00 – Communication Networks, LE/EECS 4215 3.00 – Mobile Communications, LE/EECS 4404 3.00 – Introduction to Machine Learning and Pattern Recognition, etc.
Other Desired Qualifications: Other qualifications include good communication skills.
Professor: Shahin Kamali
Contact Info: kamalis@yorku.ca
Lab Website: https://www.eecs.yorku.ca/~kamalis/
Position Type: Lassonde Undergraduate Research Award (LURA); NSERC Undergraduate Student Research Award (USRA)
Open Positions: 2
Project Description: Online algorithms are concerned with making sequential decisions for an input revealed sequentially and online. Online algorithms find applications in many domains; a rich theory studies them in the worst-case scenarios under the competitive analysis framework.
In this project, we aim to study online algorithms in the presence of machine-learned predictions about the input. These predictions must be “learnable” and may be incorrect (or even adversarially generated). The objective is to design algorithms that perform well when predictions are error-free, are robust against adversarial predictions, and achieve “fairness”; when items belong to different groups or come at different times, a fair algorithm has a solution that treats items of the same quality equally.
Duties and Responsibilities: The project involves reading research papers and relevant chapters of books on the topic, summarizing and presenting articles, giving presentations, brainstorming for novel algorithmic solutions, writing efficient code for implementing algorithms and writing reports.
Desired Technical Skills: A strong set of skills in math and theoretical computer science, A range grades in the CS theory courses (e.g., EECS-3101). Strong coding skills for running experiments.
Desired Course(s): Design and Analysis of Algorithms, Advanced Data-Structures, Advanced Algorithm Design and Analysis.
Other Desired Qualifications: N/A.
Contact Info: kamalis@yorku.ca
Lab Website: https://www.eecs.yorku.ca/~kamalis/
Position Type: Lassonde Undergraduate Research Award (LURA); NSERC Undergraduate Student Research Award (USRA)
Open Positions: 2
Project Description: Online algorithms are concerned with making sequential decisions for an input revealed sequentially and online. Online algorithms find applications in many domains; a rich theory studies them in the worst-case scenarios under the competitive analysis framework.
In this project, we aim to study online algorithms in the presence of machine-learned predictions about the input. These predictions must be “learnable” and may be incorrect (or even adversarially generated). The objective is to design algorithms that perform well when predictions are error-free, are robust against adversarial predictions, and achieve “fairness”; when items belong to different groups or come at different times, a fair algorithm has a solution that treats items of the same quality equally.
Duties and Responsibilities: The project involves reading research papers and relevant chapters of books on the topic, summarizing and presenting articles, giving presentations, brainstorming for novel algorithmic solutions, writing efficient code for implementing algorithms and writing reports.
Desired Technical Skills: A strong set of skills in math and theoretical computer science, A range grades in the CS theory courses (e.g., EECS-3101). Strong coding skills for running experiments.
Desired Course(s): Design and Analysis of Algorithms, Advanced Data-Structures, Advanced Algorithm Design and Analysis.
Other Desired Qualifications: N/A.
Professor: Michael Jenkin
Contact Info: jenkin@yorku.ca
Lab Website: https://vgr.lab.yorku.ca
Position Type: Lassonde Undergraduate Research Award (LURA); NSERC Undergraduate Student Research Award (USRA)
Open Positions: 4
Project Description: There is a need in remote areas of the world to be able to deploy Unmanned Aerial Vehicles (UAV) from Unmanned Surface Vessels (USV) to support shore-based activities. Technically, this requires solutions to a number of problems related to autonomous systems including planning for autonomous vessels and deployment recovery of aerial vehicles (typically quadcopters) from the vessel under a variety of different environmental conditions.
Duties and Responsibilities: This project involves software and hardware work related to this task. Utilizing existing USV and UAV hardware in the lab, and using Stong Pond as a proxy for a remote location in Canada, this project involves software development to support vehicle operation (path planning, localization, navigation), hardware to support UAV deployment and recovery, and landing at identified locations on land.
Desired Technical Skills: Python programming skills. Interest/knowledge of ROS 2 desirable, Interest in interfacing actuators and sensors with autonomous systems.
Desired Course(s): Undergraduate degree in Computer Science, Computer Engineering, or Software Engineering.
Other Desired Qualifications: Ability to work independently. Ability to work as part of a team.
Contact Info: jenkin@yorku.ca
Lab Website: https://vgr.lab.yorku.ca
Position Type: Lassonde Undergraduate Research Award (LURA); NSERC Undergraduate Student Research Award (USRA)
Open Positions: 4
Project Description: There is a need in remote areas of the world to be able to deploy Unmanned Aerial Vehicles (UAV) from Unmanned Surface Vessels (USV) to support shore-based activities. Technically, this requires solutions to a number of problems related to autonomous systems including planning for autonomous vessels and deployment recovery of aerial vehicles (typically quadcopters) from the vessel under a variety of different environmental conditions.
Duties and Responsibilities: This project involves software and hardware work related to this task. Utilizing existing USV and UAV hardware in the lab, and using Stong Pond as a proxy for a remote location in Canada, this project involves software development to support vehicle operation (path planning, localization, navigation), hardware to support UAV deployment and recovery, and landing at identified locations on land.
Desired Technical Skills: Python programming skills. Interest/knowledge of ROS 2 desirable, Interest in interfacing actuators and sensors with autonomous systems.
Desired Course(s): Undergraduate degree in Computer Science, Computer Engineering, or Software Engineering.
Other Desired Qualifications: Ability to work independently. Ability to work as part of a team.
Professor: Michael Jenkin
Contact Info: jenkin@yorku.ca
Lab Website: https://vgr.lab.yorku.ca
Position Type: Lassonde Undergraduate Research Award (LURA); NSERC Undergraduate Student Research Award (USRA)
Open Positions: 1
Project Description: Avatars, both mechanical and electronic, are beginning to be deployed in a range of different domains, from hotel receptionists to tour guides in museums. This project involves assisting in the development of an avatar that is designed to be deployed in commercial spaces. Leveraging existing rendering and audio processing pipelines, the avatar relies on a LLM-based engine to respond to interactions. This project will involve developing software modules to support the Avatar. Of particular interest this summer is to tune the avatar’s responses based on the perceived emotional state of the individual interacting with the avatar.
Duties and Responsibilities: Software development (primarily in Python). Interaction with industrial partners. Assistance in the deployment of the avatar in various locations in the GTA.
Desired Technical Skills: Knowledge of Python.
Desired Course(s): Undergraduate degree in Computer Science, Computer Engineering, or Software Engineering.
Other Desired Qualifications: Ability to work independently. Ability to work as part of a team. Ability to interact with external partners.
Contact Info: jenkin@yorku.ca
Lab Website: https://vgr.lab.yorku.ca
Position Type: Lassonde Undergraduate Research Award (LURA); NSERC Undergraduate Student Research Award (USRA)
Open Positions: 1
Project Description: Avatars, both mechanical and electronic, are beginning to be deployed in a range of different domains, from hotel receptionists to tour guides in museums. This project involves assisting in the development of an avatar that is designed to be deployed in commercial spaces. Leveraging existing rendering and audio processing pipelines, the avatar relies on a LLM-based engine to respond to interactions. This project will involve developing software modules to support the Avatar. Of particular interest this summer is to tune the avatar’s responses based on the perceived emotional state of the individual interacting with the avatar.
Duties and Responsibilities: Software development (primarily in Python). Interaction with industrial partners. Assistance in the deployment of the avatar in various locations in the GTA.
Desired Technical Skills: Knowledge of Python.
Desired Course(s): Undergraduate degree in Computer Science, Computer Engineering, or Software Engineering.
Other Desired Qualifications: Ability to work independently. Ability to work as part of a team. Ability to interact with external partners.
Professor: Michael Jenkin
Contact Info: jenkin@yorku.ca
Lab Website: https://vgr.lab.yorku.ca
Position Type: Lassonde Undergraduate Research Award (LURA); NSERC Undergraduate Student Research Award (USRA)
Open Positions: 1
Project Description: Working in collaboration with artist David Perrett (www.davidperrett.ca), this project will explore the use of autonomous robots as art. Building on the Boids concept, first introduced in the late 1980’s, this project will develop a robot playground to help design and deploy groups of autonomous robots that interact with visitors, the environment and each other to bring artificial life to some futuristic urban environment. In support of planned future art installations, you will lay the foundation for the behaviors and interactions for the autonomous robots and testing the reactions of the creatures in simulation. This project is related to the Artificial Urban Life project, which seeks to build prototype sample elements for this robot playground.
Duties and Responsibilities: Software development in Unity. Development of software tools to support multiple Boids operating in a simulated space.
Desired Technical Skills: C# background. Interest in developing tools in unity.
Desired Course(s): Undergraduate degree in Computer Science, Computer Engineering, Software Engineering, or Digital Media.
Other Desired Qualifications: Interest in art, interest in autonomous systems.
Contact Info: jenkin@yorku.ca
Lab Website: https://vgr.lab.yorku.ca
Position Type: Lassonde Undergraduate Research Award (LURA); NSERC Undergraduate Student Research Award (USRA)
Open Positions: 1
Project Description: Working in collaboration with artist David Perrett (www.davidperrett.ca), this project will explore the use of autonomous robots as art. Building on the Boids concept, first introduced in the late 1980’s, this project will develop a robot playground to help design and deploy groups of autonomous robots that interact with visitors, the environment and each other to bring artificial life to some futuristic urban environment. In support of planned future art installations, you will lay the foundation for the behaviors and interactions for the autonomous robots and testing the reactions of the creatures in simulation. This project is related to the Artificial Urban Life project, which seeks to build prototype sample elements for this robot playground.
Duties and Responsibilities: Software development in Unity. Development of software tools to support multiple Boids operating in a simulated space.
Desired Technical Skills: C# background. Interest in developing tools in unity.
Desired Course(s): Undergraduate degree in Computer Science, Computer Engineering, Software Engineering, or Digital Media.
Other Desired Qualifications: Interest in art, interest in autonomous systems.
Professor: Michael Jenkin
Contact Info: jenkin@yorku.ca
Lab Website: https://vgr.lab.yorku.ca
Position Type: Lassonde Undergraduate Research Award (LURA); NSERC Undergraduate Student Research Award (USRA)
Open Positions: 1
Project Description: Working in collaboration with artist David Perrett (www.davidperrett.ca), this project explores possible embodiments of artificial urban life that might occupy some future urban landscape. Much like hermit crabs will occupy trash as an external shell, might future artificial life occupy the debris and detritus of our modern world? The goal of this project is to build two instances of this futuristic world and to develop the necessary sensing and action modules to enable the occupants of this future world to interact with visitor to this game park of the future. This project is related to the robots as an artistic display project, which seeks to build a software environment to plan the interactive nature of these robots.
Duties and Responsibilities: Hardware/software design of mobile agents. Integration with large-scale simulation of multiple robots. Interest in art installations.
Desired Technical Skills: Low-level programming (python, C, others) for Raspberry Pi-based autonomous systems.
Desired Course(s): Undergraduate degree in Computer Science, Computer Engineering, Software Engineering, or Digital Media.
Other Desired Qualifications: Interest in art.
Contact Info: jenkin@yorku.ca
Lab Website: https://vgr.lab.yorku.ca
Position Type: Lassonde Undergraduate Research Award (LURA); NSERC Undergraduate Student Research Award (USRA)
Open Positions: 1
Project Description: Working in collaboration with artist David Perrett (www.davidperrett.ca), this project explores possible embodiments of artificial urban life that might occupy some future urban landscape. Much like hermit crabs will occupy trash as an external shell, might future artificial life occupy the debris and detritus of our modern world? The goal of this project is to build two instances of this futuristic world and to develop the necessary sensing and action modules to enable the occupants of this future world to interact with visitor to this game park of the future. This project is related to the robots as an artistic display project, which seeks to build a software environment to plan the interactive nature of these robots.
Duties and Responsibilities: Hardware/software design of mobile agents. Integration with large-scale simulation of multiple robots. Interest in art installations.
Desired Technical Skills: Low-level programming (python, C, others) for Raspberry Pi-based autonomous systems.
Desired Course(s): Undergraduate degree in Computer Science, Computer Engineering, Software Engineering, or Digital Media.
Other Desired Qualifications: Interest in art.
Professor: John Lam
Contact Info: johnlam@yorku.ca
Lab Website: https://pelser.lab.yorku.ca/
Position Type: Lassonde Undergraduate Research Award (LURA); NSERC Undergraduate Student Research Award (USRA)
Open Positions: 1
Project Description: As of 2022, the Information and Communication Technology (ICT) data centers’ electricity consumption in Canada has already reached more than 100 peta-joules. To ensure that data centers’ energy demand is met with minimal environmental impact, it is essential to shift towards green power solutions (such as PV energy) to power data centers. Since most of the electrical components and equipment in a data center (e.g., batteries for providing back-up power) require DC power, high voltage DC power distribution (~380V-400V) provides an attractive power architecture for data center applications to achieve high power conversion efficiency. Grid-supported PV power conversion for use in DC-distributed systems requires two separate power interfaces: a DC/DC converter that interfaces with the PV energy source and an AC/DC converter that interfaces with the grid.
This research project is to develop a single-stage GaN (Gallium Nitride) power converter module with double-input ports that can interface with the PV power source and the grid simultaneously for powering the data centers. Due to the inherent fast switching characteristics of the GaN devices, it enables very high frequency power conversion to be feasible in the designed converter topology. The developed double-input converter will be capable of (1) providing step-up/down voltage conversion to match the voltage level (i.e. to 380V) in the high voltage DC distribution network; (2) providing PV maximum power point tracking function for a wide range of PV irradiation levels and (3) minimizing the AC grid-side input current harmonics. The developed power circuit topology will be analyzed and its performance will be verified using power electronics simulation software, such as PowerSIM. Preliminary hardware validation on a proof-of-concept prototype will also be performed at the Advanced Power Electronics Laboratory for Sustainable Energy Research (PELSER).
Since this research project will facilitate the development of a new power converter that increases the efficiency and economic viability of grid-supported PV energy systems for powering data centers, the development and adoption of more efficient PV-powered data centers will reduce Greenhouse Gas (GHG) emissions significantly, and result in environmental and health benefits to Canadians.
Duties and Responsibilities: Weekly meeting with project supervisor. Perform power converter circuit analysis, and verify design concept in Power Electronics simulation. Perform preliminary hardware validation of the developed research idea.
Desired Technical Skills: Problem solving skills, electronic circuit analysis, solid mathematical skills, programming skills in MATLAB.
Desired Course(s): LE/EECS 2200 3.00 – Electrical Circuits, LE/EECS 3611 4.00 – Analog Electronics, LE/EECS 4613 4.00 – Power Electronics, LE/EECS 3201 4.00 – Digital Logic Design, LE/EECS 4623 4.00 – Renewable Energy Systems.
Other Desired Qualifications: Hardworking, teamwork, strong written and oral communication skills.
Contact Info: johnlam@yorku.ca
Lab Website: https://pelser.lab.yorku.ca/
Position Type: Lassonde Undergraduate Research Award (LURA); NSERC Undergraduate Student Research Award (USRA)
Open Positions: 1
Project Description: As of 2022, the Information and Communication Technology (ICT) data centers’ electricity consumption in Canada has already reached more than 100 peta-joules. To ensure that data centers’ energy demand is met with minimal environmental impact, it is essential to shift towards green power solutions (such as PV energy) to power data centers. Since most of the electrical components and equipment in a data center (e.g., batteries for providing back-up power) require DC power, high voltage DC power distribution (~380V-400V) provides an attractive power architecture for data center applications to achieve high power conversion efficiency. Grid-supported PV power conversion for use in DC-distributed systems requires two separate power interfaces: a DC/DC converter that interfaces with the PV energy source and an AC/DC converter that interfaces with the grid.
This research project is to develop a single-stage GaN (Gallium Nitride) power converter module with double-input ports that can interface with the PV power source and the grid simultaneously for powering the data centers. Due to the inherent fast switching characteristics of the GaN devices, it enables very high frequency power conversion to be feasible in the designed converter topology. The developed double-input converter will be capable of (1) providing step-up/down voltage conversion to match the voltage level (i.e. to 380V) in the high voltage DC distribution network; (2) providing PV maximum power point tracking function for a wide range of PV irradiation levels and (3) minimizing the AC grid-side input current harmonics. The developed power circuit topology will be analyzed and its performance will be verified using power electronics simulation software, such as PowerSIM. Preliminary hardware validation on a proof-of-concept prototype will also be performed at the Advanced Power Electronics Laboratory for Sustainable Energy Research (PELSER).
Since this research project will facilitate the development of a new power converter that increases the efficiency and economic viability of grid-supported PV energy systems for powering data centers, the development and adoption of more efficient PV-powered data centers will reduce Greenhouse Gas (GHG) emissions significantly, and result in environmental and health benefits to Canadians.
Duties and Responsibilities: Weekly meeting with project supervisor. Perform power converter circuit analysis, and verify design concept in Power Electronics simulation. Perform preliminary hardware validation of the developed research idea.
Desired Technical Skills: Problem solving skills, electronic circuit analysis, solid mathematical skills, programming skills in MATLAB.
Desired Course(s): LE/EECS 2200 3.00 – Electrical Circuits, LE/EECS 3611 4.00 – Analog Electronics, LE/EECS 4613 4.00 – Power Electronics, LE/EECS 3201 4.00 – Digital Logic Design, LE/EECS 4623 4.00 – Renewable Energy Systems.
Other Desired Qualifications: Hardworking, teamwork, strong written and oral communication skills.
Professor: Hany Farag
Contact Info: hefarag@yorku.ca
Lab Website: https://smartgrid.eecs.yorku.ca
Position Type: Lassonde Undergraduate Research Award (LURA); NSERC Undergraduate Student Research Award (USRA)
Open Positions: 2
Project Description: Proliferation of low carbon hydrogen-based technologies as a means for alleviation of the dependence to fossil fuel energies has become a priority for many entities across the globe. Deployment of various clean hydrogen plants is imperative to pave the way for materialization and promotion of various hydrogen-based technologies. Based on the Hydrogen Strategy unveiled in December 2020, Canada aims to position itself as a world-leading producer, user, and exporter of clean hydrogen. In response, Ontario has developed its provisional strategy for a low-carbon hydrogen economy, which was published in April 2022. Ontario has planned out multiple immediate actions to rapidly increase production capacity of low-carbon hydrogen to meet its need. Recently, Canada and Germany signed an agreement to enhance German energy security with clean Canadian hydrogen. The Alliance will build on Canada’s Hydrogen Strategy and mark a step toward Canada’s objective of becoming a top global supplier of clean hydrogen.
Hydrogen can be produced using an electrolyzer, known as a Power-to-Gas (PtG) technology, whereby electricity is utilized to diffuse water into hydrogen and oxygen. Also, Steam Methane Reforming (SMR) can utilize methane gas as a feedstock to produce hydrogen, where methane can be accessed from the natural gas grid. Methane can be also produced from renewable biogas resources, e.g., landfills, agricultural residues, waste, etc. The generated hydrogen from electrolysis and/or SMR can be stored in the form of liquid or gas in a reservoir for later use. Hydrogen can then be converted back into electricity and supplied to the power grid using a fuel cell and/or Gas-fired Generators (GfG), or it can be directly sold to the transportation sector (i.e., to supply FCEVs) and other applications such as refineries, steel production, and agriculture. Without appropriate regional and/or provincial-wide technical and economic studies for the mass adoption of EHPs and evaluation of their impacts on the electricity system and hydrogen demand, creation of governments’ policies and investment plans might reach a point of diminishing returns.
In this project, the research team will explore innovative approaches and develop engineering tools for modeling, design, and performance optimization of MW-scale electrolysis hydrogen plants from renewable energy. The developed tools will be utilized as a decision-making support toolbox to evaluate and enhance the technical and economic aspects associated with the wide adoption of hydrogen plants in the province of Ontario with consideration of its integration with the electricity system to: (1) facilitate seamless integration for higher penetration levels of renewable, yet intermittent, energy resources such as wind and solar, (2) provide flexible operation options for the Independent Electricity System Operator (IESO) via using hydrogen plants as fast response and long term energy storage technologies, and (3) makes hydrogen production affordable enough to compete with its counterparts.
Duties and Responsibilities: Work with graduate students and postdocs in the research team in the following tasks:
A. Develop Region-specific hydrogen demand uptake scenarios: (i) Identify and allocate sectors/industries for hydrogen uptake in the province: e.g., hard-to-abate sectors (refineries, chemical industry, steelmaking), transportation, power generation, and heating. (ii) Identify key import/export corridors for outside-province hydrogen transport. (iii) Using existing statistically relevant data (transportation surveys, FCEV sales data, academic journals, consulting agencies reporting, building stock and projected access to hydrogen) interpolate hydrogen demand growth projections performing a regression analysis to derive annual rates of hydrogen demand uptake (high, med, low) in the IESO regions based on factors that are determined to be relevant. (iv) Consider different hydrogen demand uptake projections within different regions/municipalities concerns relating to geographical considerations, mode share targets to encourage FCEV-based transit, and co-location of green hydrogen generation and utilization e.g., hydraulic in Niagara, and nuclear in GTA East. (v) Reporting on where hydrogen demand exists and at which rates and times. (vi) Derive the hourly/daily/weekly hydrogen demand profile in each region. (vii) Convert the hydrogen demand in ton to its equivalent electric energy in MWh. (viii) Develop heat maps of modeled electricity/hydrogen demand uptake in each region for years 5, 10, and 15. (ix) Reporting, including source files and calculations. Tables and supporting data estimating 10-year growth of hydrogen demand.
B. Develop Region-specific green hydrogen production scenarios: (i) Generate different scenarios for allocation and sizing of EHPs to meet the projected hydrogen demand: (ia) Off-site and co-located with renewable power generators e.g., hydraulic, nuclear, wind. (ib) Off-site with minimum distance to hydrogen demand. (ic) Off-site with least negative impacts on the electricity system e.g., system losses, transmission congestions, etc. (id) Distributed on-site hydrogen generation (e.g., bus depot, steelmaking factory). (ie) Develop province-wide heat maps for hydrogen production scenarios. (ii) Estimate the annual electricity costs of green hydrogen production for each facility with consideration of: (iia) Class A versus Class B hydrogen producers. (iib) Participation in the existing ICI program. (iic) Participation in other programs such as CDR. (iid) Introduce different mechanisms for “interruptible pricing rate” to replace/improve the existing ICI program. (iie) Identify the optimal “interruptible pricing rate” that minimizes the electricity costs of hydrogen producers without negatively impacting other customers and/or ratepayers. (iii) Estimate the capacity factor and costs of EHPs under different application objectives. (iiia) EHPs could be oversized and operate at low CF. (iiib) EHPs could be sized to operate at fixed rated power and/or supply base hydrogen demand. (iiic) EHPs could be sized to follow the intermittency of variable generation.
C. Develop Green hydrogen storage and transportation scenarios: (i) Generate different scenarios for green hydrogen storage (daily, weekly, seasonal) considering their impacts on. (ia) Elasticity of green hydrogen production and its impact on the flexibility of the electricity grid (without reasonably sized storage, hydrogen produces may not be able to control their electricity demand and provide grid services). (ib) Complexity and costs of hydrogen transportation. (ic) Reliability of hydrogen supply. (id) Costs of hydrogen production.
Desired Technical Skills: Coding in MatLab, strong in statistical analysis, writing in Latex, and preferred to have knowledge in optimization.
Desired Course(s): A/A+ in third/fourth year electric power engineering courses, preferred to have basic knowledge about electrolysis, renewable energy resources, and energy storage.
Other Desired Qualifications: Good communication skills, passionate about climate change, willing to be mentored by graduate students and/or postdocs, and willing to be available in-person (in the smart grid lab) most of the weekdays.
Contact Info: hefarag@yorku.ca
Lab Website: https://smartgrid.eecs.yorku.ca
Position Type: Lassonde Undergraduate Research Award (LURA); NSERC Undergraduate Student Research Award (USRA)
Open Positions: 2
Project Description: Proliferation of low carbon hydrogen-based technologies as a means for alleviation of the dependence to fossil fuel energies has become a priority for many entities across the globe. Deployment of various clean hydrogen plants is imperative to pave the way for materialization and promotion of various hydrogen-based technologies. Based on the Hydrogen Strategy unveiled in December 2020, Canada aims to position itself as a world-leading producer, user, and exporter of clean hydrogen. In response, Ontario has developed its provisional strategy for a low-carbon hydrogen economy, which was published in April 2022. Ontario has planned out multiple immediate actions to rapidly increase production capacity of low-carbon hydrogen to meet its need. Recently, Canada and Germany signed an agreement to enhance German energy security with clean Canadian hydrogen. The Alliance will build on Canada’s Hydrogen Strategy and mark a step toward Canada’s objective of becoming a top global supplier of clean hydrogen.
Hydrogen can be produced using an electrolyzer, known as a Power-to-Gas (PtG) technology, whereby electricity is utilized to diffuse water into hydrogen and oxygen. Also, Steam Methane Reforming (SMR) can utilize methane gas as a feedstock to produce hydrogen, where methane can be accessed from the natural gas grid. Methane can be also produced from renewable biogas resources, e.g., landfills, agricultural residues, waste, etc. The generated hydrogen from electrolysis and/or SMR can be stored in the form of liquid or gas in a reservoir for later use. Hydrogen can then be converted back into electricity and supplied to the power grid using a fuel cell and/or Gas-fired Generators (GfG), or it can be directly sold to the transportation sector (i.e., to supply FCEVs) and other applications such as refineries, steel production, and agriculture. Without appropriate regional and/or provincial-wide technical and economic studies for the mass adoption of EHPs and evaluation of their impacts on the electricity system and hydrogen demand, creation of governments’ policies and investment plans might reach a point of diminishing returns.
In this project, the research team will explore innovative approaches and develop engineering tools for modeling, design, and performance optimization of MW-scale electrolysis hydrogen plants from renewable energy. The developed tools will be utilized as a decision-making support toolbox to evaluate and enhance the technical and economic aspects associated with the wide adoption of hydrogen plants in the province of Ontario with consideration of its integration with the electricity system to: (1) facilitate seamless integration for higher penetration levels of renewable, yet intermittent, energy resources such as wind and solar, (2) provide flexible operation options for the Independent Electricity System Operator (IESO) via using hydrogen plants as fast response and long term energy storage technologies, and (3) makes hydrogen production affordable enough to compete with its counterparts.
Duties and Responsibilities: Work with graduate students and postdocs in the research team in the following tasks:
A. Develop Region-specific hydrogen demand uptake scenarios: (i) Identify and allocate sectors/industries for hydrogen uptake in the province: e.g., hard-to-abate sectors (refineries, chemical industry, steelmaking), transportation, power generation, and heating. (ii) Identify key import/export corridors for outside-province hydrogen transport. (iii) Using existing statistically relevant data (transportation surveys, FCEV sales data, academic journals, consulting agencies reporting, building stock and projected access to hydrogen) interpolate hydrogen demand growth projections performing a regression analysis to derive annual rates of hydrogen demand uptake (high, med, low) in the IESO regions based on factors that are determined to be relevant. (iv) Consider different hydrogen demand uptake projections within different regions/municipalities concerns relating to geographical considerations, mode share targets to encourage FCEV-based transit, and co-location of green hydrogen generation and utilization e.g., hydraulic in Niagara, and nuclear in GTA East. (v) Reporting on where hydrogen demand exists and at which rates and times. (vi) Derive the hourly/daily/weekly hydrogen demand profile in each region. (vii) Convert the hydrogen demand in ton to its equivalent electric energy in MWh. (viii) Develop heat maps of modeled electricity/hydrogen demand uptake in each region for years 5, 10, and 15. (ix) Reporting, including source files and calculations. Tables and supporting data estimating 10-year growth of hydrogen demand.
B. Develop Region-specific green hydrogen production scenarios: (i) Generate different scenarios for allocation and sizing of EHPs to meet the projected hydrogen demand: (ia) Off-site and co-located with renewable power generators e.g., hydraulic, nuclear, wind. (ib) Off-site with minimum distance to hydrogen demand. (ic) Off-site with least negative impacts on the electricity system e.g., system losses, transmission congestions, etc. (id) Distributed on-site hydrogen generation (e.g., bus depot, steelmaking factory). (ie) Develop province-wide heat maps for hydrogen production scenarios. (ii) Estimate the annual electricity costs of green hydrogen production for each facility with consideration of: (iia) Class A versus Class B hydrogen producers. (iib) Participation in the existing ICI program. (iic) Participation in other programs such as CDR. (iid) Introduce different mechanisms for “interruptible pricing rate” to replace/improve the existing ICI program. (iie) Identify the optimal “interruptible pricing rate” that minimizes the electricity costs of hydrogen producers without negatively impacting other customers and/or ratepayers. (iii) Estimate the capacity factor and costs of EHPs under different application objectives. (iiia) EHPs could be oversized and operate at low CF. (iiib) EHPs could be sized to operate at fixed rated power and/or supply base hydrogen demand. (iiic) EHPs could be sized to follow the intermittency of variable generation.
C. Develop Green hydrogen storage and transportation scenarios: (i) Generate different scenarios for green hydrogen storage (daily, weekly, seasonal) considering their impacts on. (ia) Elasticity of green hydrogen production and its impact on the flexibility of the electricity grid (without reasonably sized storage, hydrogen produces may not be able to control their electricity demand and provide grid services). (ib) Complexity and costs of hydrogen transportation. (ic) Reliability of hydrogen supply. (id) Costs of hydrogen production.
Desired Technical Skills: Coding in MatLab, strong in statistical analysis, writing in Latex, and preferred to have knowledge in optimization.
Desired Course(s): A/A+ in third/fourth year electric power engineering courses, preferred to have basic knowledge about electrolysis, renewable energy resources, and energy storage.
Other Desired Qualifications: Good communication skills, passionate about climate change, willing to be mentored by graduate students and/or postdocs, and willing to be available in-person (in the smart grid lab) most of the weekdays.