Deep Reinforcement Learning
The goal of this webbook is to keep track of the state-of-the-art in deep reinforcement learning. It starts with basics in reinforcement learning and deep learning to introduce the notations. It then covers different classes of deep RL methods, value-based or policy-based, model-free or model-based, etc. Later sections focus on more advanced topics.
This document is meant to stay work in progress forever, as new algorithms will be added as they are published. Feel free to comment, correct, suggest, pull request by writing to julien.vitay@gmail.com.
Some figures are taken from the original publication (“Source:” in the caption). Their copyright stays to the respective authors, naturally. The rest is my own work and can be distributed, reproduced and modified under CC-BY-SA-NC 4.0.
Except where otherwise noted, this work is licensed under a Creative Commons Attribution-Non Commercial-ShareAlike 4.0 International License.