Fall 2015 Course Descriptions
In this semester the course will focus on agents that learn, plan and act in non-deterministic and complex environments. We will first cover the main relevant results and approaches from Markov Decision Processes and Reinforcement Learning (RL) and then turn to recent research papers. Advanced topics include: combining planning and learning, factored representations, relational RL, hierarchical RL, partial programming, predictive state representations, approximation methods, convergence guarantees.
Prerequisite: Math 22, Comp 15, Comp 160 or consent of instructor. Some exposure to probability theory is necessary. Previous courses in machine learning or artificial intelligence are helpful but not required.