Overview

This repository provides code, exercises and solutions for popular Reinforcement Learning algorithms. These are meant to serve as a learning tool to complement the theoretical materials from

• Reinforcement Learning: An Introduction (2nd Edition)
• David Silver's Reinforcement Learning Course

Each folder in corresponds to one or more chapters of the above textbook and/or course. In addition to exercises and solution, each folder also contains a list of learning goals, a brief concept summary, and links to the relevant readings.

All code is written in Python 3 and uses RL environments from OpenAI Gym. Advanced techniques use Tensorflow for neural network implementations.

Table of Contents

• Introduction to RL problems & OpenAI Gym
• MDPs and Bellman Equations
• Dynamic Programming: Model-Based RL, Policy Iteration and Value Iteration
• Monte Carlo Model-Free Prediction & Control
• Temporal Difference Model-Free Prediction & Control
• Function Approximation
• Deep Q Learning (WIP)
• Policy Gradient Methods (WIP)
• Learning and Planning (WIP)
• Exploration and Exploitation (WIP)

List of Implemented Algorithms

• [Dynamic Programming Policy Evaluation](DP/Policy Evaluation Solution.ipynb)
• [Dynamic Programming Policy Iteration](DP/Policy Iteration Solution.ipynb)
• [Dynamic Programming Value Iteration](DP/Value Iteration Solution.ipynb)
• [Monte Carlo Prediction](MC/MC Prediction Solution.ipynb)
• [Monte Carlo Control with Epsilon-Greedy Policies](MC/MC Control with Epsilon-Greedy Policies Solution.ipynb)
• [Monte Carlo Off-Policy Control with Importance Sampling](MC/Off-Policy MC Control with Weighted Importance Sampling Solution.ipynb)
• [SARSA (On Policy TD Learning)](TD/SARSA Solution.ipynb)
• [Q-Learning (Off Policy TD Learning)](TD/Q-Learning Solution.ipynb)
• [Q-Learning with Linear Function Approximation](FA/Q-Learning with Value Function Approximation Solution.ipynb)
• [Deep Q-Learning for Atari Games](DQN/Deep Q Learning Solution.ipynb)
• [Double Deep-Q Learning for Atari Games](DQN/Double DQN Solution.ipynb)
• Deep Q-Learning with Prioritized Experience Replay (WIP)
• [Policy Gradient: REINFORCE with Baseline](PolicyGradient/CliffWalk REINFORCE with Baseline Solution.ipynb)
• [Policy Gradient: Actor Critic with Baseline](PolicyGradient/CliffWalk Actor Critic Solution.ipynb)
• [Policy Gradient: Actor Critic with Baseline for Continuous Action Spaces](PolicyGradient/Continuous MountainCar Actor Critic Solution.ipynb)
• Deterministic Policy Gradients for Continuous Action Spaces (WIP)
• Deep Deterministic Policy Gradients (DDPG) (WIP)
• Asynchronous Advantage Actor Critic (A3C)

Resources

Textbooks:

• Reinforcement Learning: An Introduction (2nd Edition)

Classes:

• David Silver's Reinforcement Learning Course (UCL, 2015)
• CS294 - Deep Reinforcement Learning (Berkeley, Fall 2015)
• CS 8803 - Reinforcement Learning (Georgia Tech)

Talks/Tutorials:

• Introduction to Reinforcement Learning (Joelle Pineau @ Deep Learning Summer School 2016)
• Deep Reinforcement Learning (Pieter Abbeel @ Deep Learning Summer School 2016)
• Deep Reinforcement Learning ICML 2016 Tutorial (David Silver)
• Tutorial: Introduction to Reinforcement Learning with Function Approximation
• John Schulman - Deep Reinforcement Learning (4 Lectures)
• Deep Reinforcement Learning Slides @ NIPS 2016

Other Projects:

• carpedm20/deep-rl-tensorflow
• matthiasplappert/keras-rl

Selected Papers:

• Human-Level Control through Deep Reinforcement Learning (2015-02)
• Deep Reinforcement Learning with Double Q-learning (2015-09)
• Continuous control with deep reinforcement learning (2015-09)
• Prioritized Experience Replay (2015-11)
• Dueling Network Architectures for Deep Reinforcement Learning (2015-11)
• Asynchronous Methods for Deep Reinforcement Learning (2016-02)
• Deep Reinforcement Learning from Self-Play in Imperfect-Information Games (2016-03)
• Mastering the game of Go with deep neural networks and tree search