CMSC389F at University of Maryland

# Reinforcement Learning

Lectures: F 12:00-12:50 p.m., 3118 Csic

## Instructor Kevin Chen

kev (at) umd.edu

## Instructor Zack Khan

zack123 (at) umd.edu

### Week 1 Overview

## Introduction to Reinforcement Learning

### Week 2 Overview

## Reinforcement Learning Framework and Markov Decision Processes

### Week 3 Overview

## Markov Decision Processes With Gridworld

### Week 4 Overview

## Discounting and Cumulative Reward

### Week 6 Overview

## DP Methods: Value and Policy Iteration

### Week 9 Overview

## Temporal Difference Learning

## Problem Sets

It is highly-recommended that you complete the problem-sets. Late submissions are deducted 10% the following day, and any later submissions are not accepted. See Syllabus for more information.

- Problem Set 01: Reinforcement Learning Basics ( TeX)
- Problem Set 02: Python and Colaboratory
- Problem Set 03: Decisions and MDPs
- Problem Set 04: Discounting
- Problem Set 05: Value Functions
- Problem Set 06:Value and Policy Iteration
- Problem Set 07: Open AI Gym and Random Policy

Expand

## Lecture Slides

Slides generally follow the notes on a weekly basis. See Syllabus for more information.

- Slides 01: Reinforcement Learning
- Slides 02: Markov Decision Processes
- Slides 06: Value and Policy Iteration
- Slides 07: Monte Carlo
- Slides 08: TD Learning
- Slides 09: TD Lambda + Q Learning Intro
- Slides 10: Q Learning

Expand