CS 498 RL1

Subject offerings of new and developing areas of knowledge in computer science intended to augment the existing curriculum. See Class Schedule or departmental course information for topics and prerequisites. Course Information: 1 to 4 undergraduate hours. 1 to 4 graduate hours. May be repeated in the same or separate terms if topics vary.

Reinforcement learning (RL) is a machine learning paradigm for sequential decision-making, which has enabled the recent successes in video/board game playing (e.g., AlphaGo). In this course we will introduce the fundamental concepts and some basic algorithms for RL. Most of the course will be highly mathematical, and the goal is to enable students to (1) understand the mathematical framework of RL, (2) tell what problems can be solved with RL, and how to express these problems using the RL formulation, (3) understand why and how RL algorithms are designed to work in theory, and (4) know how to experimentally and mathematically evaluate the effectiveness of an RL algorithm. There will be both programming and written assignments. Prerequisites: Required: Linear algebra (Math 415 or equivalent), Probability and Statistics (CS 361 or equivalent). Recommended: Numerical methods (CS 357 or 450), AI or Machine Learning (CS 440 and/or 446). For up-to-date information about CS course restricti