During the Spring 2018 the University of Waterloo is offering a graduate course on Reinforcement Learning.

CS885 - Reinforcement Learning (Spring 2018)

The course introduces students to the design of algorithms that enable machines to learn based on reinforcements. Applications of reinforcement learning include robotic control, autonomous vehicles, game playing, conversational agents, assistive technologies, computational finance, operations research, etc.

Class 1a - Introduction and Markov Processes

Class 1b - Markov Processes

Class 2a - Markov Decision Processes

Class 2b - Value Iteration

Class 3a - Policy Iteration

Class 3b - Introduction to RL

Class 4a - Deep Neural Networks

Class 4b - Deep Q-Networks

Class 5 - Conversational Agents

Class 6a - OpenAI Environments

Class 6b - DQN and TesorFlow

Class 7a - Policy Gradient

Class 7b - Actor Critic

Class 8a - Multi-armed Bandits

Class 8b - Bayesian and Contextual Bandits

Class 9 - Model-based RL

Class 10 - Bayesian RL

Class 11a - Hidden Markov Models

Class 11b - Partially Observable RL

Class 12 - Deep Recurrent Q-Networks

Class 13a - Playing FPS Games with Deep RL

Class 13b - Lifelong Learning in Minecraft

Class 13c - Adversarial Search

Class 14a - Mastering the Game of Go

Class 14b - Mastering Chess and Shogi

Class 14c - Trust Region Models

Class 15a - Trust Region Policy Optimization

Class 15b - Proximal Policy Optimization

Class 15c - Semi-Markov Decision Processes

Class 16a - The Option-Critic Architecture

Class 16b - FeUdal Networks for Hierarchical RL

Class 17a - Target-Driven Visual Navigation

Class 17b - Control of a Quadrotor

Class 17c - Inverse Reinforcement Learning

Class 18a - Safe Multi-Agent RL for Autonomous Driving

Class 18b: Learning Driving Styles for Autonomous Vehicles

Class 19a - End-To-End LSTM Based Dialog Control

Class 19b - Learning Cooperative Visual Dialog Agents

Class 19c - Memory Augmented Networks

Class 20a - Neural Map: Structured Memory for Deep RL

Class 20b - Memory Augmented Control Networks