• List of Reinforcement Learning 전반: Reinforcement Learning Algorithms in Python,OpenAI Gym, Tensorflow [2,185 stars on Github]

Article / post

• Simple reinforcement learning methods to learn CartPole : OpenAI gym에서 만나는 가장 쉬운 강화학습 샘플인 카트폴을 상세히 풀이하여 소개
• 강화학습 튜토리알 - 인공 신경망으로 ‘퐁’ 게임을 학습시키자 : Karpathy의 Pong 강화학습 튜토리얼을 한글로 훌륭하게 번역
• Simple Reinforcement Learning : 텐서플로우를 이용한 강화학습 자료를 올리고 있는 Arthur Julinani의 미디엄
• Machine Learning 스터디 (20) Reinforcement Learning : 스탠포드 CS229 강의 노트를 기반으로 작성한 강화학습 한글 문서
• Simple Beginner’s guide to Reinforcement Learning & its implementation
• Deep Learning Research Review: Reinforcement Learning : By Adit Deshpande, UCLA.
• Youtube:Tutorial: Introduction to Reinforcement Learning with Function Approximation,[ppt]
• 이웅권의 Reinforement GitBook
• Deep Reinforcement Learning (John Schulman, OpenAI) :[Youtube1] and [Youtube2]
• 추천_모음
• eBook: Reinforcement Learning: An Introduction
• ICML 2016 RL by David Silver :일반적인 RL,알파고 집중 분석
• Kim Sung's RL Study plan
• 모두의연구소_Reinforcement Learning 그리고 OpenAI
• 강화학습 튜토리알 - 인공 신경망으로 '퐁' 게임을 학습시키자 (Andrej Karpathy 포스트 번역)_
• ppt_InfoGAN: Interpretable Representation Learning by Information Maximizing Generative Adversarial Nets
• 강화 학습 기초 Reinforcement Learning an introduction : 김태훈(carpedm20)
• ICML: Tutorial Deep Reinforcement Learning : David Silver

• Torch DQN 강화학습 소개

• 카카오AI 리포트-강화학습의 비밀

• The Next Challenges for Reinforcement Learning

• Evolution Strategies as a Scalable Alternative to Reinforcement Learning: 진화알고리즘 적용, OpenAI

• Introduction to Q-Learning

• 강화학습 예제 14: RLCode팀이 직접 만든 강화학습 예제들을 모아놓은 Repo 입니다

• Asynchronous Advantage Actor-Critic (A3C), [분석글], [논문]

• DQN의 단점을 보안한 제안(2016)

• Episode 1 — Genetic Algorithm for Reinforcement Learning

• one-step-Q-learning algorithm for this AI : q-learning 을 이용한 틱-택-토 AI 를 만들어보았습니다.

Youtube

• 알아두면 쓸데있는 신기한 강화학습: 김태훈, ​발표 슬라이드

• RLCode와 A3C 쉽고 깊게 이해하기

Tutorial

• GitHub: Practical_RL : HSE대학교 강의 자료

• Gitbook:Fundamental of Reinforcement Learning: 모두의 연구소 Woong won, Lee

• 강화학습 그리고 OpenAI: 모두의 연구실 자료

• 1. Introduction to OpenAI
• 2-1. Intro to Reinforcement Learning (1) MDP & Value Function
• 2-2. Intro to Reinforcement Learning (2) Q Learning
• 3-1. CartPole with Deep Q Learning (1) CartPole example
• 3-2. CartPole with Deep Q Learning (2) DQN
• 3-3. CartPole with Deep Q Learning (3) TensorFlow
• 3-4. CartPole with Deep Q Learning (4) Code review
• 4-1. CartPole with Policy Gadient (1) Policy Gradient
• 4-2. CartPole with Policy Gadient (2) Code review
• 5-1. about Atari games
• 5-2. Atari "Pong"
• 5-3. Atari "Breakout"

• 유투브:RL Course by David Silver - Lecture 1: Introduction to Reinforcement Learning

• MuJoCo

Paper

• Deep Reinforcement Learning: An Overview : Yuxi Li

• FeUdal Networks for Hierarchical Reinforcement Learning : Alexander Sasha Vezhnevets, Simon Osindero, Tom Schaul, Nicolas Heess, Max Jaderberg, David Silver, Koray Kavukcuoglu(2017.03)

• Overcoming catastrophic forgetting in neural networks: 강화학습 + transfer-learning

• the concept of transfering the learning of reinforcement agents from environment to environment.

Inverse reinforcement-learning

• ppt: Inverse Reinforcement Learning Algorithms : Sungjoon Samuel 작성

VizDoom

• 설명_설치: 3D 환경에서 동작하는 Agent

Implementation

• A set of Deep Reinforcement Learning Agents implemented in Tensorflow. : Reinforcement Learning with TensorFlow&OpenAI Gym강의

• 텐서플로우(TensorFlow)를 이용해 간단한 DQN(Deep-Q-Networks) 에이전트를 만들어 보기 (CatchGame)

• Human-Level Control through Deep Reinforcement Learning: Songrotek, DQN-Tensorflow

• Deep Reinforcement Learning for Keras

4. Simple Reinforcement Learning with Tensorflow

• Part 0: Q-Learning with Tables and Neural Networks, [번역]
• Part 1 - Two-armed Bandit,[번역]
• Part 1.5: Contextual Bandits, [번역]
• Part 2 : Policy-Based Agents, [번역]
• Part 3 : Model-Based RL, [번역]
• Part 4 : Deep Q-Networks and Beyond, [번역]
• Part 5 : Visualizing an Agent’s Thoughts and Actions, [번역]
• Part 6 : Partial Observability and Deep Recurrent Q-Networks, [번역]
• Part 7 : Action-Selection Strategies for Exploration, [번역]
• Part 8 : Asynchronous Actor-Critic Agents (A3C), [번역]

Learning Reinforcement Learning (with Code, Exercises and Solutions) : 6장 구성, 추천

• 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)

2.

Lecture 1: 수업의 개요 비디오 강의 슬라이드 Lecture 2: OpenAI GYM 게임해보기 비디오 강의 슬라이드

Lab 2: OpenAI GYM 게임해보기 실습 비디오 실습슬라이드

Lecture 3: Dummy Q-learning (table) 비디오 강의 슬라이드

Lab 3: Dummy Q-learning (table) 비디오 실습슬라이드

Lecture 4: Q-learning exploit&exploration and discounted reward 비디오 강의 슬라이드

Lab 4: Q-learning exploit&exploration and discounted reward 비디오 실습슬라이드

Lecture 5: Q-learning in non-deterministic world 비디오 강의 슬라이드

Lab 5: Q-learning in non-deterministic world 비디오 실습슬라이드 Lab 5-1: Q-learning web Demo 비디오

Lecture 6: Q-Network 비디오 강의 슬라이드

Lab 6-1: Q Network for Frozen Lake 비디오 실습슬라이드 Lab 6-2: Q Network for Cart Pole 비디오 실습슬라이드

Lecture 7: DQN 비디오 강의 슬라이드

Lab 7-1: DQN 1 (NIPS 2013) 비디오 실습슬라이드 Lab 7-2: DQN 2 (Nature 2015) 비디오 실습슬라이드 Lab 7-3: DQN Cart Pole Demo 비디오 Lab 7-4: DQN Simple Pacman Demo (여러분은 최고 몇점까지 갈수 있나요?) 비디오

ICML 2016 Tutorials

• All list

Material (ppt, pdf)

• [eBook]Reinforcement Learning:An Introduction: Richard S. Sutton and Andrew G. Barto, 2017, MIT출판, 538p

implementation

• Reinforcement Agent to play the Helicopter Game