![A tv screen displays virtual soccer players in the midst of a soccer game.](https://brocku.ca/brock-news/wp-content/uploads/2022/07/Virtual-soccer-game-1600x1066.jpg?x91846)
NOTE: This is the latest in a series of question-and-answer stories featuring faculty members who are integrating the Niagara 2022 Canada Games into the courses they teach at Brock University or the research they’re leading. For more information on Brock’s academic activities around the Games, visit brocku.ca/canada-games
Yifeng Li is an Assistant Professor of Computer Science in the Faculty of Mathematics and Science. His research and teaching interests include artificial intelligence (AI) and machine learning; data science and data analytics; optimization and computational intelligence; and computational biology and bioinformatics.
![A man in glasses sits in front of a grey wall while wearing a navy blue sweater.](https://brocku.ca/brock-news/wp-content/uploads/2022/07/Yifeng-Li.png?x91846)
Assistant Professor of Computer Science Yifeng Li.
What is your Canada Games-related course title, code and description?
COSC 4P83/5P83 Reinforcement Learning.
This course introduces fundamental concepts and applications of reinforcement learning (RL) algorithms — the core family of AI techniques for sequential decision-making tasks. An RL computer agent is trained to take the optimal action that maximizes its long-term accumulated rewards for a situational state. The course covers basic concepts, such as dynamic programming, Monte Carlo methods and temporal-difference algorithms, as well as neural network-based RL methods — deep reinforcement learning (DRL) models, to learn value functions, such as expected accumulated rewards if the agent starts from a given state, and policies, such as action taking modules. For the applications of RL, this course focuses on training computer agent players for various computer games.
Describe how you’ve integrated Canada Games related material into your course?
This course requires each student to apply the AI algorithms they learned to train computer agents in games, such as tennis, hockey and soccer. Each student needs to present their idea, including sport type and algorithms to be used, in the middle of the term; make the final presentation to show their results including a video clip of how these computer agents behave in games; and submit a short report to summarize the project. Even though the prototypic projects developed in this course won’t be directly used by Canada Games, the course motivates its students and instructor to draw inspiration from them. We have also heard that some of our students are volunteering at the Games, and some graduate students are applying RL in their thesis projects.
Why do you think the Canada Games presents such a good opportunity for students at Brock?
The Canada Games provides a unique opportunity to allow students to perceive the power of their AI knowledge learned in the classroom and transfer it to meaningful applications. It motivates students to pursue their continuous learning in AI and sports.
Do you have any suggestions for ways your colleagues can use the Games to enhance teaching and learning opportunities in their courses?
We should optimize our teaching and students’ success from a multi-dimensional perspective. Embedding Games-related topics in teaching examples, exercises, assignments and projects can help students better understand the class materials and real-world values and connections.
Once the Games are finished, how do you plan to continue using this new idea in your course?
I will continue to use the Games as a focus in the final projects of this course and look for more resources to improve students’ experiences.
For more information about COSC 4P83/5P83, please contact Li at yli2@brocku.ca