The mechanics of motivation: Can we teach an agent to play Pac-Man?

Every reinforcement learning model tells a story about the incentives it acts under. Explore the foundational mechanics of intelligent agents, the design of dynamic reward systems, and the core stages of reinforcement learning: from environment setup and trial-and-error learning to policy optimization. Learn how reward formulation shapes agent behavior and discover the practical frameworks essential for building your own Q-learning systems. Leave with a portfolio artifact that teaches an agent to play a game.

This event follows the new format for AI Club workshops this year. You will commence a year-long journey to collect project artifacts and build your very own portfolio for AI, data science, and ML roles.