ECE524: Foundations of Reinforcement Learning
Spring 2024, Tue. Thu. 09:30 am - 10:50 am, Equad B205,
Instructor: Chi Jin Office hour: Fri. 3:00-4:00 pm, Equad C332
TA: Wenhao Zhan Office hour: Mon. 3:00-4:00 pm, Friend Center 308.
Contents: Mathematical foundations of RL, mostly about theorems and proofs.
Grades: 5 problem sets (60%), 1 final exam (40%).
No late homework.
Problem Sets
Schedule (weekly basis)
Basics (tabular MDP):
Intro, MDP basics and planning.
Concentration inequalities.
Generative models, value iteration.
Online RL, exploration, optimism. [Homework 1 due]
Minimax lower bound.
Offline RL, pessimism. [Homework 2 due]
Advanced Topics:
Policy optimization.
Large state space, linear function approximation. [Homework 3 due]
General function approximation.
Game theory and multiagent RL . [Homework 4 due]
Learning Markov games.
Partial observable MDP. [Homework 5 due]
Reference Readings
Reinforcement Learning: Theory and Algorithms (draft), by Alekh Agarwal, Nan Jiang, Sham M. Kakade, Wen Sun
Reinforcement learning: an introduction, by Richard S. Sutton, Andrew G. Barto
Algorithms for Reinforcement Learning, by Csaba Szepesvári
Bandit Algorithms, by Tor Lattimore, Csaba Szepesvari
Mathematical Tools
High dimensional probability. An introduction with applications in Data Science, by Roman Vershynin
Concentration inequalities and martingale inequalities — a survey, by Fan Chung, Linyuan Lu
Related Courses
Nan Jiang, Statistical Reinforcement Learning
Wen Sun and Sham Kakade, Foundations of Reinforcement Learning
Dylan J. Foster and Alexander Rakhlin, Statistical Reinforcement Learning and Decision Making
Alekh Agarwal and Alex Slivkins, Bandits and Reinforcement Learning
More practical/empirical version (will not be covered in this course)
Sergey Levine, Deep Reinforcement Learning