Reinforcement Learning[论文合集]

https://handong1587.github.io/deep_learning/2015/10/09/rl.html

---

Jump to...

     1. Tutorials

    1. Simple Reinforcement Learning with Tensorflow

     2.  Courses

     3.  Papers

    1. Surveys

    2. ​​​​​​Playing Doom

     4.  Projects

    1. ​​​​​​​Autonomous vehicle navigation

    2. ​​​​​​​Play Flappy Bird

     5.  Pong

    1. ​​​​​​​Tips and Tricks

     6. Library

     7. Blogs

             1. ​​​​Let’s make a DQN

     8. Books

     9. Resources

     10. Reading and Questions

---

Tutorials

Demystifying Deep Reinforcement Learning (Part1)

http://neuro.cs.ut.ee/demystifying-deep-reinforcement-learning/

Deep Reinforcement Learning With Neon (Part2)

http://neuro.cs.ut.ee/deep-reinforcement-learning-with-neon/

Deep Reinforcement Learning

• intro: David Silver, Google DeepMind
• slides: http://www.iclr.cc/lib/exe/fetch.php?media=iclr2015:silver-iclr2015.pdf
• mirror: http://pan.baidu.com/s/1qWBOJGo

Deep Reinforcement Learning

• intro: MLSS 2016. John Schulman[UC Berkeley]
• homepage: http://rl-gym-doc.s3-website-us-west-2.amazonaws.com/mlss/index.html
• slides: http://pan.baidu.com/s/1jIatusA#path=%252F

Deep Reinforcement Learning: Pong from Pixels

• intro: Andrej Karpathy
• blog: http://karpathy.github.io/2016/05/31/rl/
• gist: https://gist.github.com/karpathy/a4166c7fe253700972fcbc77e4ea32c5

Deep Reinforcement Learning

• instructor: David Silver. RLDM 2015
• video: http://videolectures.net/rldm2015_silver_reinforcement_learning/

Deep Reinforcement Learning

• intro: David Silver [Google DeepMind]
• video: http://techtalks.tv/talks/deep-reinforcement-learning/62360/
• slides: http://hunch.net/~beygel/deep_rl_tutorial.pdf

The Nuts and Bolts of Deep RL Research

• intro: NIPS 2016, John Schulman, OpenAI
• slides: http://rll.berkeley.edu/deeprlcourse/docs/nuts-and-bolts.pdf
• mirror: https://pan.baidu.com/s/1kVkBLkF

ML Tutorial: Modern Reinforcement Learning and Video Games

• intro: by Marc Bellemare [DeepMind]
• youtube: https://www.youtube.com/watch?v=WuFMrk3ZbkE
• mirror: https://www.bilibili.com/video/av17360035/

Reinforcement learning explained

• blog: https://www.oreilly.com/ideas/reinforcement-learning-explained

Beginner’s guide to Reinforcement Learning & its implementation in Python

https://www.analyticsvidhya.com/blog/2017/01/introduction-to-reinforcement-learning-implementation/

Reinforcement Learning on the Web

• intro: Andrej Karpathy
• slides: https://docs.google.com/presentation/d/1lcYrN56V2_SuX1rSmpzOUeMnheF6Jsu33-MsvLW9O_4/edit#slide=id.p
• slides: http://alpha.openai.com/ak_rework_2017.pdf

Deep Q Learning with Keras and Gym

• blog: https://keon.io/rl/deep-q-learning-with-keras-and-gym/
• github: https://github.com/keon/deep-q-learning

“Deep Reinforcement Learning, Decision Making, and Control

• intro: ICML 2017 Tutorial
• slides: https://sites.google.com/view/icml17deeprl

A Tour of Reinforcement Learning: The View from Continuous Control

• intro: by Benjamin Recht, UC Berkeley
• slides: https://people.eecs.berkeley.edu/~brecht/l2c-icml2018/Recht_ICML_Control-RL_tutorial.pdf

An Introduction to Deep Reinforcement Learning

• intro: McGill University & Google Brain
• arxiv: https://arxiv.org/abs/1811.12560

Simple Reinforcement Learning with Tensorflow

Part 0: Q-Learning with Tables and Neural Networks https://medium.com/emergent-future/simple-reinforcement-learning-with-tensorflow-part-0-q-learning-with-tables-and-neural-networks-d195264329d0#.oo105wa2t

Part 1 - Two-armed Bandit

https://medium.com/@awjuliani/super-simple-reinforcement-learning-tutorial-part-1-fd544fab149#.tk89k51ob

Part 2 - Policy-based Agents

https://medium.com/@awjuliani/super-simple-reinforcement-learning-tutorial-part-2-ded33892c724#.n2wytg9q0

Part 3 - Model-Based RL https://medium.com/@awjuliani/simple-reinforcement-learning-with-tensorflow-part-3-model-based-rl-9a6fe0cce99#.742i2yj6p

Part 4: Deep Q-Networks and Beyond https://medium.com/@awjuliani/simple-reinforcement-learning-with-tensorflow-part-4-deep-q-networks-and-beyond-8438a3e2b8df#.jox069crz

Part 5: Visualizing an Agent’s Thoughts and Actionshttps://medium.com/@awjuliani/simple-reinforcement-learning-with-tensorflow-part-5-visualizing-an-agents-thoughts-and-actions-4f27b134bb2a#.pluh6cygm

Part 6: Partial Observability and Deep Recurrent Q-Networks

• blog: https://medium.com/emergent-future/simple-reinforcement-learning-with-tensorflow-part-6-partial-observability-and-deep-recurrent-q-68463e9aeefc#.3se46qkzy
• github: https://gist.github.com/awjuliani/35d2ab3409fc818011b6519f0f1629df

Part 7: Action-Selection Strategies for Exploration

• blog: https://medium.com/emergent-future/simple-reinforcement-learning-with-tensorflow-part-7-action-selection-strategies-for-exploration-d3a97b7cceaf#.8mcaa5nbe
• demo: https://awjuliani.github.io/exploration/index.html

Dissecting Reinforcement Learning

• part 1: https://mpatacchiola.github.io/blog/2016/12/09/dissecting-reinforcement-learning.html
• part 2: https://mpatacchiola.github.io/blog/2017/01/15/dissecting-reinforcement-learning-2.html
• part 3: https://mpatacchiola.github.io/blog/2017/01/29/dissecting-reinforcement-learning-3.html
• github: https://github.com/mpatacchiola/dissecting-reinforcement-learning

REINFORCE tutorial

• intro: A small collection of code snippets and notes explaining the foundations of the REINFORCE algorithm.
• github: https://github.com/mathias-madsen/reinforce_tutorial

Deep Q-Learning Recap

http://blog.davidqiu.com/Research/%5B%20Recap%20%5D%20Deep%20Q-Learning%20Recap/

Introduction to Reinforcement Learning

• intro: Joelle Pineau [McGill University]
• video: http://videolectures.net/deeplearning2016_pineau_reinforcement_learning/
• slides: http://videolectures.net/site/normal_dl/tag=1051677/deeplearning2016_pineau_reinforcement_learning_01.pdf

Courses

Advanced Topics: RL

UCL Course on RL

• instructors: David Silver (Google DeepMind, AlphaGo)
• homepage: http://www0.cs.ucl.ac.uk/staff/d.silver/web/Teaching.html
• youtube: https://www.youtube.com/playlist?list=PLqYmG7hTraZDM-OYHWgPebj2MfCFzFObQ
• video: http://pan.baidu.com/s/1bnWGuIz/
• assignment: http://www0.cs.ucl.ac.uk/staff/d.silver/web/Teaching_files/Easy21-Johannes.pdf

CS 294: Deep Reinforcement Learning, Fall 2017

• instructor: Sergey Levine
• homepage: http://rll.berkeley.edu/deeprlcourse/
• youtube: https://www.youtube.com/playlist?list=PLkFD6_40KJIznC9CDbVTjAF2oyt8_VAe3
• bilibili: https://www.bilibili.com/video/av21501169/

CS 294: Deep Reinforcement Learning, Spring 2017

• course page: http://rll.berkeley.edu/deeprlcoursesp17/
• github: https://github.com//txizzle/drl

Berkeley CS 294: Deep Reinforcement Learning

• instructors: John Schulman, Pieter Abbeel
• homepage: http://rll.berkeley.edu/deeprlcourse/
• youtube: https://www.youtube.com/playlist?list=PLkFD6_40KJIwTmSbCv9OVJB3YaO4sFwkX
• mirror: https://pan.baidu.com/s/1hsQcm1Y

(Udacity) Reinforcement Learning - Offered at Georgia Tech as CS 8803

• instructor: Charles Isbell, Michael Littman
• homepage: https://www.udacity.com/course/reinforcement-learning–ud600
• homepage: https://classroom.udacity.com/courses/ud820/lessons/684808907/concepts/6512308530923

CS229 Lecture notes Part XIII: Reinforcement Learning and Control

• intro: Andrew Ng
• lecture notes: http://cs229.stanford.edu/notes/cs229-notes12.pdf

Practical_RL: A course in reinforcement learning in the wild

• github: https://github.com/yandexdataschool/Practical_RL

Reinforcement Learning (COMP-762) Winter 2017

• course page: http://www.cs.mcgill.ca/~dprecup/courses/rl.html
• lectures: http://www.cs.mcgill.ca/~dprecup/courses/RL/lectures.html

**Deep RL Bootcamp - 26-27 August 2017

Berkeley CA**

• lectures: https://sites.google.com/view/deep-rl-bootcamp/lectures
• video: https://www.bilibili.com/video/av15568836/

CMPUT 366: Intelligent Systems and CMPUT 609: Reinforcement Learning & Artificial Intelligence

• intro: by Rich Sutton, Adam White
• lecture video: https://drive.google.com/drive/folders/0B3w765rOKuKAMG9lbmRacFdsLWM?direction=a

Deep Reinforcement Learning and Control (Spring 2017, CMU 10703)

• instructors: Katerina Fragkiadaki, Ruslan Satakhutdinov
• homepage: https://katefvision.github.io/
• video: https://www.youtube.com/playlist?list=PLpIxOj-HnDsNPFdu2UqCu2McJKHs-eWXv
• mirror: https://www.bilibili.com/video/av18865689/

Advanced Deep Learning & Reinforcement Learning

• intro: DeepMind
• youtube: https://www.youtube.com/playlist?list=PLqYmG7hTraZDNJre23vqCGIVpfZ_K2RZs
• bilibili: https://www.bilibili.com/video/av36621866/
• github: https://github.com/RylanSchaeffer/ucl-adv-dl-rl

Papers

Playing Atari with Deep Reinforcement Learning

• intro: Google DeepMind. NIPS Deep Learning Workshop 2013
• arxiv: http://arxiv.org/abs/1312.5602
• github: https://github.com/kristjankorjus/Replicating-DeepMind
• demo: http://cs.stanford.edu/people/karpathy/convnetjs/demo/rldemo.html
• github: https://github.com/Kaixhin/Atari
• github(Tensorflow): https://github.com/gliese581gg/DQN_tensorflow
• summary: https://github.com/aleju/papers/blob/master/neural-nets/Playing_Atari_with_Deep_Reinforcement_Learning.md

Deep Learning for Real-Time Atari Game Play Using Offline Monte-Carlo Tree Search Planning

• intro: NIPS 2014
• keywords: DQN, MCTS
• paper: http://papers.nips.cc/paper/5421-scalable-inference-for-neuronal-connectivity-from-calcium-imaging
• paper: https://web.eecs.umich.edu/~baveja/Papers/UCTtoCNNsAtariGames-FinalVersion.pdf

Replicating the Paper “Playing Atari with Deep Reinforcement Learning”

• intro: University of Tartu
• technical report: https://courses.cs.ut.ee/MTAT.03.291/2014_spring/uploads/Main/Replicating%20DeepMind.pdf

A Tutorial for Reinforcement Learning

• paper: http://web.mst.edu/~gosavia/tutorial.pdf
• code(C): http://web.mst.edu/~gosavia/bookcodes.html
• code(Matlab): http://web.mst.edu/~gosavia/mrrl_website.html

Incentivizing Exploration In Reinforcement Learning With Deep Predictive Models

• arxiv: http://arxiv.org/abs/1507.00814
• notes: https://www.evernote.com/shard/s189/sh/a4262b84-a322-4f77-9a76-569278be84af/b8c3e146a76ca3853f560bb03b60a481

Massively Parallel Methods for Deep Reinforcement Learning

• intro: ICML 2015. DeepMind
• keywords: DQN, Gorila
• arxiv: https://arxiv.org/abs/1507.04296

Action-Conditional Video Prediction using Deep Networks in Atari Games

• homepage: https://sites.google.com/a/umich.edu/junhyuk-oh/action-conditional-video-prediction
• arxiv: http://arxiv.org/abs/1507.08750
• github: https://github.com/junhyukoh/nips2015-action-conditional-video-prediction
• video: http://video.weibo.com/show?fid=1034:98062f3d83e41da6faa99cde5aa1ac97

Deep Recurrent Q-Learning for Partially Observable MDPs

• intro: AAAI 2015
• arxiv: https://arxiv.org/abs/1507.06527

Continuous control with deep reinforcement learning

• intro: Google DeepMind
• arxiv: http://arxiv.org/abs/1509.02971
• github: https://github.com/iassael/torch-policy-gradient
• github: https://github.com/stevenpjg/ddpg-aigym
• github(TensorFlow + OpenAI Gym): https://github.com/SimonRamstedt/ddpg

Benchmarking for Bayesian Reinforcement Learning

• arxiv: http://arxiv.org/abs/1509.04064
• code: https://github.com/mcastron/BBRL/
• reading: http://blogs.ulg.ac.be/damien-ernst/benchmarking-for-bayesian-reinforcement-learning/

Deep Reinforcement Learning with Double Q-learning

• intro: AAAI 2016
• arxiv: https://arxiv.org/abs/1509.06461

Giraffe: Using Deep Reinforcement Learning to Play Chess

• arxiv: http://arxiv.org/abs/1509.01549

Human-level control through deep reinforcement learning

• intro: Google DeepMind. 2015 Nature
• paper: http://www.readcube.com/articles/10.1038/nature14236?shared_access_token=Lo_2hFdW4MuqEcF3CVBZm9RgN0jAjWel9jnR3ZoTv0P5kedCCNjz3FJ2FhQCgXkApOr3ZSsJAldp-tw3IWgTseRnLpAc9xQq-vTA2Z5Ji9lg16_WvCy4SaOgpK5XXA6ecqo8d8J7l4EJsdjwai53GqKt-7JuioG0r3iV67MQIro74l6IxvmcVNKBgOwiMGi8U0izJStLpmQp6Vmi_8Lw_A%3D%3D
• paper: http://web.stanford.edu/class/psych209/Readings/MnihEtAlHassibis15NatureControlDeepRL.pdf
• github(Lua/Torch): https://github.com/deepmind/dqn
• mirror: http://pan.baidu.com/s/1kTiwzOF
• code: https://sites.google.com/a/deepmind.com/dqn/
• youtube: https://www.youtube.com/watch?v=V2wzkPmiB_A
• github: https://github.com/kuz/DeepMind-Atari-Deep-Q-Learner
• github: https://github.com/tambetm/simple_dqn
• github: https://github.com/devsisters/DQN-tensorflow
• reddit: https://www.reddit.com/r/MachineLearning/comments/2x4yy1/google_deepmind_nature_paper_humanlevel_control

Data-Efficient Learning of Feedback Policies from Image Pixels using Deep Dynamical Models

• arxiv: http://arxiv.org/abs/1510.02173

Variational Information Maximisation for Intrinsically Motivated Reinforcement Learning

• intro: Google DeepMind
• arxiv: http://arxiv.org/abs/1509.08731
• notes: https://www.evernote.com/shard/s189/sh/8c7ff9d9-c321-4e83-a802-58f55ebed9ac/bfc614113180a5f4624390df56e73889

Actor-Mimic: Deep Multitask and Transfer Reinforcement Learning

• intro: ICLR 2016
• arxiv: http://arxiv.org/abs/1511.06342
• github: https://github.com/eparisotto/ActorMimic

MazeBase: A Sandbox for Learning from Games

• intro: New York University & Facebook AI Research
• arxiv: http://arxiv.org/abs/1511.07401

Learning Simple Algorithms from Examples

• intro: New York University & Facebook AI Research
• arxiv: http://arxiv.org/abs/1511.07275
• github: https://github.com/wojzaremba/algorithm-learning

Learning Algorithms from Data

• PhD thesis: http://www.cs.nyu.edu/media/publications/zaremba_wojciech.pdf
• github: https://github.com/wojzaremba/algorithm-learning

Multiagent Cooperation and Competition with Deep Reinforcement Learning

• arxiv: http://arxiv.org/abs/1511.08779
• github: https://github.com/NeuroCSUT/DeepMind-Atari-Deep-Q-Learner-2Player

Active Object Localization with Deep Reinforcement Learning

• arxiv: http://arxiv.org/abs/1511.06015

Deep Reinforcement Learning with Attention for Slate Markov Decision Processes with High-Dimensional States and Actions

• arxiv: http://arxiv.org/abs/1512.01124

How to Discount Deep Reinforcement Learning: Towards New Dynamic Strategies

• arxiv: http://arxiv.org/abs/1512.02011

State of the Art Control of Atari Games Using Shallow Reinforcement Learning

• arxiv: http://arxiv.org/abs/1512.01563

Angrier Birds: Bayesian reinforcement learning

• arxiv: http://arxiv.org/abs/1601.01297
• github: https://github.com/imanolarrieta/angrybirds
• gitxiv: http://gitxiv.com/posts/Nr2N7j4YrR4gnCYK9/angrier-birds-bayesian-reinforcement-learning

Prioritized Experience Replay

• arxiv: http://arxiv.org/abs/1511.05952

Dueling Network Architectures for Deep Reinforcement Learning

• intro: ICML 2016 best paper
• arxiv: http://arxiv.org/abs/1511.06581
• notes: https://hadovanhasselt.wordpress.com/2016/06/20/best-paper-at-icml-dueling-network-architectures-for-deep-reinforcement-learning/

Asynchronous Methods for Deep Reinforcement Learning

• arxiv: http://arxiv.org/abs/1602.01783
• github(Tensorflow): https://github.com/traai/async-deep-rl
• github(Tensorflow+Keras+OpenAI Gym): https://github.com/coreylynch/async-rl
• github(Tensorflow): https://github.com/devsisters/async-rl-tensorflow
• github(PyTorch): https://github.com/ikostrikov/pytorch-a3c
• notes: https://blog.acolyer.org/2016/10/10/asynchronous-methods-for-deep-reinforcement-learning/

Graying the black box: Understanding DQNs

• arxiv: http://arxiv.org/abs/1602.02658

Learning to Communicate to Solve Riddles with Deep Distributed Recurrent Q-Networks

• arxiv: http://arxiv.org/abs/1602.02672

Value Iteration Networks

• intro: NIPS 2016, Best Paper Award. University of California, Berkeley
• arxiv: http://arxiv.org/abs/1602.02867
• github(official, Theano): https://github.com/avivt/VIN
• github: https://github.com/TheAbhiKumar/tensorflow-value-iteration-networks
• github: https://github.com/onlytailei/PyTorch-value-iteration-networks
• github: https://github.com/kentsommer/pytorch-value-iteration-networks
• github: https://github.com/neka-nat/vin-keras
• notes(by Andrej Karpathy): https://github.com/karpathy/paper-notes/blob/master/vin.md

Insights in Reinforcement Learning

• intro: MSc thesis
• mirror: http://pan.baidu.com/s/1bn51BYJ

Using Deep Q-Learning to Control Optimization Hyperparameters

• arxiv: http://arxiv.org/abs/1602.04062

Continuous Deep Q-Learning with Model-based Acceleration

• arxiv: http://arxiv.org/abs/1603.00748

Deep Reinforcement Learning from Self-Play in Imperfect-Information Games

• arxiv: http://arxiv.org/abs/1603.01121

Hierarchical Deep Reinforcement Learning: Integrating Temporal Abstraction and Intrinsic Motivation

• intro: MIT
• arxiv: https://arxiv.org/abs/1604.06057
• github: https://github.com/EthanMacdonald/h-DQN

Benchmarking Deep Reinforcement Learning for Continuous Control

• arxiv: http://arxiv.org/abs/1604.06778
• github: https://github.com/rllab/rllab
• doc: https://rllab.readthedocs.org/en/latest/

Terrain-Adaptive Locomotion Skills Using Deep Reinforcement Learning

• homepage: http://www.cs.ubc.ca/~van/papers/2016-TOG-deepRL/index.html
• paper: http://www.cs.ubc.ca/~van/papers/2016-TOG-deepRL/2016-TOG-deepRL.pdf
• github: https://github.com/xbpeng/DeepTerrainRL

Hierarchical Reinforcement Learning using Spatio-Temporal Abstractions and Deep Neural Networks

• arxiv: http://arxiv.org/abs/1605.05359

Deep Successor Reinforcement Learning (MIT)

• arxiv: http://arxiv.org/abs/1606.02396
• github: https://github.com/Ardavans/DSR

Learning to Communicate with Deep Multi-Agent Reinforcement Learning

• arxiv: https://arxiv.org/abs/1605.06676
• github: https://github.com/iassael/learning-to-communicate

Deep Reinforcement Learning with Regularized Convolutional Neural Fitted Q Iteration RC-NFQ: Regularized Convolutional Neural Fitted Q Iteration

• intro: A batch algorithm for deep reinforcement learning. Incorporates dropout regularization and convolutional neural networks with a separate target Q network.
• paper: http://machineintelligence.org/papers/rc-nfq.pdf
• github: https://github.com/cosmoharrigan/rc-nfq

Episodic Exploration for Deep Deterministic Policies: An Application to StarCraft Micromanagement Tasks

• intro: Facebook AI Research
• arxiv: http://arxiv.org/abs/1609.02993

Bayesian Reinforcement Learning: A Survey

• arxiv: http://arxiv.org/abs/1609.04436

Playing FPS Games with Deep Reinforcement Learning

• arxiv: http://arxiv.org/abs/1609.05521
• demo: https://www.youtube.com/playlist?list=PLduGZax9wmiHg-XPFSgqGg8PEAV51q1FT
• notes: https://blog.acolyer.org/2016/11/23/playing-fps-games-with-deep-reinforcement-learning/

Reset-Free Guided Policy Search: Efficient Deep Reinforcement Learning with Stochastic Initial States

• intro: University of Washington & UC Berkeley
• arxiv: https://arxiv.org/abs/1610.01112

Utilization of Deep Reinforcement Learning for saccadic-based object visual search

• arxiv: https://arxiv.org/abs/1610.06492

Learning to Navigate in Complex Environments

• intro: Google DeepMind
• arxiv: https://arxiv.org/abs/1611.03673
• github: https://github.com/deepmind/lab
• youtube: https://www.youtube.com/watch?v=lNoaTyMZsWI

Reinforcement Learning with Unsupervised Auxiliary Tasks

• intro: DeepMind. ICLR 2017 oral
• arxiv: https://arxiv.org/abs/1611.05397

Learning to reinforcement learn

• intro: DeepMind
• arxiv: https://arxiv.org/abs/1611.05763

A Deep Learning Approach for Joint Video Frame and Reward Prediction in Atari Games

• intro: Graduate Training Center of Neuroscience & MSR
• arxiv: https://arxiv.org/abs/1611.07078

Exploration for Multi-task Reinforcement Learning with Deep Generative Models

• intro: NIPS Deep Reinforcement Learning Workshop 2016
• arxiv: https://arxiv.org/abs/1611.09894

Neural Combinatorial Optimization with Reinforcement Learning

• intro: Google Brain
• keywords: traveling salesman problem (TSP)
• arxiv: https://arxiv.org/abs/1611.09940

Loss is its own Reward: Self-Supervision for Reinforcement Learning

• arxiv: https://arxiv.org/abs/1612.07307

Reinforcement Learning Using Quantum Boltzmann Machines

• intro: 1QB Information Technologies (1QBit)
• arxiv: https://arxiv.org/abs/1612.05695

Deep Reinforcement Learning applied to the game Bubble Shooter

• bachelor thesis: https://staff.fnwi.uva.nl/b.bredeweg/pdf/BSc/20152016/Samson.pdf
• github: https://github.com/laurenssam/AlphaBubble
• demo: https://www.youtube.com/watch?v=DPAKFenNgbs

Deep Reinforcement Learning: An Overview

• arxiv: https://arxiv.org/abs/1701.07274

Robust Adversarial Reinforcement Learning

• intro: CMU & Google Brain & Google Research
• arxiv: https://arxiv.org/abs/1703.02702

Beating Atari with Natural Language Guided Reinforcement Learning

• intro: Stanford University
• arxiv: https://arxiv.org/abs/1704.05539

Feature Control as Intrinsic Motivation for Hierarchical Reinforcement Learning

• intro: Imperial College London
• arxiv: https://arxiv.org/abs/1705.06769
• github: https://github.com/Nat-D/FeatureControlHRL

Distral: Robust Multitask Reinforcement Learning

• intro: DeepMind
• keywords: Distill, transfer learning
• arxiv: https://arxiv.org/abs/1707.04175

Deep Reinforcement Learning: Framework, Applications, and Embedded Implementations

• intro: Syracuse University & University of California, Riverside
• arxiv: https://arxiv.org/abs/1710.03792

Robust Deep Reinforcement Learning with Adversarial Attacks

https://arxiv.org/abs/1712.03632

Variational Deep Q Network

• intro: Second workshop on Bayesian Deep Learning (NIPS 2017). Columbia University
• arxiv: https://arxiv.org/abs/1711.11225

On Monte Carlo Tree Search and Reinforcement Learning

https://www.jair.org/media/5507/live-5507-10333-jair.pdf

Distributed Deep Reinforcement Learning: Learn how to play Atari games in 21 minutes

• intro: deepsense.ai & Intel & Polish Academy of Sciences
• arxiv: https://arxiv.org/abs/1801.02852
• gihtub: https://github.com//anonymous-author1/DDRL

GAN Q-learning

https://arxiv.org/abs/1805.04874

Finding and Visualizing Weaknesses of Deep Reinforcement Learning Agents

• intro: Visual Geometry Group, University of Oxford & Element AI & Polytechnique Montreal, Mila & Canada CIFAR AI Chair
• arxiv: https://arxiv.org/abs/1904.01318

Surveys

Reinforcement Learning: A Survey

• intro: JAIR 1996
• project page: http://www.cs.cmu.edu/afs/cs/project/jair/pub/volume4/kaelbling96a-html/rl-survey.html
• arxiv: http://arxiv.org/abs/cs/9605103

A Brief Survey of Deep Reinforcement Learning

• intro: IEEE Signal Processing Magazine, Special Issue on Deep Learning for Image Understanding
• intro: Imperial College London & Arizona State University
• arxiv: https://arxiv.org/abs/1708.05866

Playing Doom

ViZDoom: A Doom-based AI Research Platform for Visual Reinforcement Learning

• arxiv: http://arxiv.org/abs/1605.02097
• github: https://github.com/Marqt/ViZDoom
• homepage: http://vizdoom.cs.put.edu.pl/
• tutorial: http://vizdoom.cs.put.edu.pl/tutorial

Deep Reinforcement Learning From Raw Pixels in Doom

• intro: Bachelor’s thesis
• arxiv: https://arxiv.org/abs/1610.02164

Playing Doom with SLAM-Augmented Deep Reinforcement Learning

• intro: University of Oxford
• arxiv: https://arxiv.org/abs/1612.00380

Reinforcement Learning via Recurrent Convolutional Neural Networks

• intro: ICPR 2016
• arxiv: https://arxiv.org/abs/1701.02392
• github: https://github.com/tanmayshankar/RCNN_MDP

Shallow Updates for Deep Reinforcement Learning

• intro: The Technion & UC Berkeley
• arxiv: https://arxiv.org/abs/1705.07461
• github(Official): https://github.com/Shallow-Updates-for-Deep-RL/Shallow_Updates_for_Deep_RL

Projects

TorchQLearning

• github: https://github.com/SeanNaren/TorchQLearningExample

General_Deep_Q_RL: General deep Q learning framework

• github: https://github.com/VinF/General_Deep_Q_RL
• wiki: https://github.com/VinF/General_Deep_Q_RL/wiki

Snake: Toy example of deep reinforcement model playing the game of snake

• github: https://github.com/bitwise-ben/Snake

Using Deep Q Networks to Learn Video Game Strategies

• github: https://github.com/asrivat1/DeepLearningVideoGames

qlearning4k: Q-learning for Keras

 

• intro: “Qlearning4k is a reinforcement learning add-on for the python deep learning library Keras. Its simple, and is ideal for rapid prototyping.”
• github: https://github.com/farizrahman4u/qlearning4k

rlenvs: Reinforcement learning environments for Torch7, inspired by RL-Glue

• github: https://github.com/Kaixhin/rlenvs

deep_rl_ale: An implementation of Deep Reinforcement Learning / Deep Q-Networks for Atari games in TensorFlow

• github: https://github.com/Jabberwockyll/deep_rl_ale

Chimp: General purpose framework for deep reinforcement learning

• github: https://github.com/sisl/Chimp

Deep Q Learning for ATARI using Tensorflow

• github: https://github.com/mrkulk/deepQN_tensorflow

DeepQLearning: A powerful machine learning algorithm utilizing Q-Learning and Neural Networks, implemented using Torch and Lua.

• github: https://github.com/blakeMilner/DeepQLearning

OpenAI Gym: A toolkit for developing and comparing reinforcement learning algorithms

• homepage: https://gym.openai.com/
• github: https://github.com/openai/gym

DeeR: DEEp Reinforcement learning framework

• github: https://github.com/VinF/deer/
• docs: http://deer.readthedocs.io/en/latest/

KeRLym: A Deep Reinforcement Learning Toolbox in Keras

• homepage: https://oshearesearch.com/index.php/2016/06/14/kerlym-a-deep-reinforcement-learning-toolbox-in-keras/
• github: https://github.com/osh/kerlym

Pack of Drones: Layered reinforcement learning for complex behaviors

• github: https://github.com/MickyDowns/deep-theano-rnn-lstm-car
• youtube: https://www.youtube.com/watch?v=WrLRGzbfeZc

RL Helicopter Game: Q-Learning and DQN Reinforcement Learning to play the Helicopter Game - Keras based!

• project page: http://dandxy89.github.io/rf_helicopter/
• github: https://github.com/dandxy89/rf_helicopter

Playing Mario with Deep Reinforcement Learning

• github: https://github.com/aleju/mario-ai

Deep Attention Recurrent Q-Network

• intro: Deep Reinforcement Learning Workshop, NIPS 2015. DeepHack Game
• arxiv: https://arxiv.org/abs/1512.01693
• github: https://github.com/5vision/DARQN

Deep Reinforcement Learning in TensorFlow

• intro: TensorFlow implementation of Deep Reinforcement Learning papers
• github: https://github.com/carpedm20/deep-rl-tensorflow

rltorch: A RL package for Torch that can also be used with openai gym

• github: https://github.com/ludc/rltorch

deep_q_rl: Theano-based implementation of Deep Q-learning

• github: https://github.com/spragunr/deep_q_rl

Reinforcement-trading

• intro: This project uses reinforcement learning on stock market and agent tries to learn trading. The goal is to check if the agent can learn to read tape. The project is dedicated to hero in life great Jesse Livermore.
• github: https://github.com/deependersingla/deep_trader

dist-dqn：Distributed Reinforcement Learning using Deep Q-Network in TensorFlow

• github: https://github.com/viswanathgs/dist-dqn

Deep Reinforcement Learning for Keras

• github: https://github.com/matthiasplappert/keras-rl

RL4J: Reinforcement Learning for the JVM

• intro: Reinforcement learning framework integrated with deeplearning4j.
• github: https://github.com/deeplearning4j/rl4j

Teaching Your Computer To Play Super Mario Bros. – A Fork of the Google DeepMind Atari Machine Learning Project

• blog: http://www.ehrenbrav.com/2016/08/teaching-your-computer-to-play-super-mario-bros-a-fork-of-the-google-deepmind-atari-machine-learning-project/
• github: https://github.com/ehrenbrav/DeepQNetwork

dprl: Deep reinforcement learning package for torch7

• github: https://github.com/PoHsunSu/dprl

Reinforcement Learning for Torch: Introducing torch-twrl

• blog: https://blog.twitter.com/2016/reinforcement-learning-for-torch-introducing-torch-twrl
• github: https://github.com/twitter/torch-twrl

Alpha Toe - Using Deep learning to master Tic-Tac-Toe - Daniel Slater

• blog: http://www.danielslater.net/2016/10/alphatoe.html
• youtube: https://www.youtube.com/watch?v=Meb5hApAnj4
• github: https://github.com/DanielSlater/AlphaToe

Tensorflow-Reinforce: Implementation of Reinforcement Learning Models in Tensorflow

• github: https://github.com/yukezhu/tensorflow-reinforce

deep RL hacking on minecraft with malmo

• github: https://github.com/matpalm/malmomo

ReinforcementLearning

• intro: MC control, Q-learning, SARSA, Cross Entropy Method
• github: https://github.com/janivanecky/ReinforcementLearning

markovjs: Reinforcement Learning in JavaScript

• github: https://github.com/lsunsi/markovjs

Deep Q: Deep reinforcement learning with TensorFlow

• github: https://github.com/tobegit3hub/deep_q

Deep Q-Learning Network in pytorch

https://github.com/transedward/pytorch-dqn

Tensorflow-RL: Implementations of deep RL papers and random experimentation

https://github.com/steveKapturowski/tensorflow-rl

Minimal and Clean Reinforcement Learning Examples

https://github.com/rlcode/reinforcement-learning

DeepRL: Highly modularized implementation of popular deep RL algorithms by PyTorch

https://github.com/ShangtongZhang/DeepRL

Autonomous vehicle navigation

Self-Driving-Car-AI

• intro: A simple self-driving car AI python script using the deep Q-learning algorithm
• github: https://github.com//JianyangZhang/Self-Driving-Car-AI

Autonomous vehicle navigation based on Deep Reinforcement Learning

https://github.com//kaihuchen/DRL-AutonomousVehicles

Car Racing using Reinforcement Learning

• intro: Stanford University
• paper: https://web.stanford.edu/class/cs221/2017/restricted/p-final/elibol/final.pdf

Play Flappy Bird

Using Deep Q-Network to Learn How To Play Flappy Bird

• github: https://github.com/yenchenlin/DeepLearningFlappyBird

Playing Flappy Bird Using Deep Reinforcement Learning (Based on Deep Q Learning DQN using Tensorflow)

• blog: http://blog.csdn.net/songrotek/article/details/50951537
• github: https://github.com/songrotek/DRL-FlappyBird

Playing Flappy Bird Using Deep Reinforcement Learning (Based on Deep Q Learning DQN)

• github: https://github.com/li-haoran/DRL-FlappyBird

MXNET-Scala Playing Flappy Bird Using Deep Reinforcement Learning

• github: https://github.com/Ldpe2G/DeepLearningForFun/tree/master/Mxnet-Scala/DRLFlappyBird

Flappy Bird Bot using Reinforcement Learning in Python

• github: https://github.com/chncyhn/flappybird-qlearning-bot

Using Keras and Deep Q-Network to Play FlappyBird

• blog: https://yanpanlau.github.io/2016/07/10/FlappyBird-Keras.html
• github: https://github.com/yanpanlau/Keras-FlappyBird

Pong

Building a Pong playing AI in just 1 hour(plus 4 days training…)

• sildes: https://speakerdeck.com/danielslater/building-a-pong-ai
• github: https://github.com/DanielSlater/PyDataLondon2016
• youtube: https://www.youtube.com/watch?v=n8NdT_3y9oY

Pong Neural Network(LIVE)

• youtube: https://www.youtube.com/watch?v=Hqf__FlRlzg
• github: https://github.com/llSourcell/pong_neural_network_live

Tips and Tricks

DeepRLHacks

• intro: The Nuts and Bolts of Deep RL Research
• github: https://github.com/williamFalcon/DeepRLHacks

Library

BURLAP: Brown-UMBC Reinforcement Learning and Planning (BURLAP) java code library

• intro: for the use and development of single or multi-agent planning and learning algorithms and domains to accompany them
• homepage: http://burlap.cs.brown.edu/

AgentNet: Deep Reinforcement Learning library for humans

• intro: A lightweight library to build and train deep reinforcement learning and custom recurrent networks using Theano+Lasagne
• github: https://github.com/yandexdataschool/AgentNet

Atari Multitask & Transfer Learning Benchmark (AMTLB)

• intro: Atari gauntlet for RL agents
• project page: http://ai-on.org/projects/multitask-and-transfer-learning.html
• github: https://github.com/deontologician/atari_multitask

Coach: a python reinforcement learning research framework containing implementation of many state-of-the-art algorithms

• intro: Reinforcement Learning Coach by Intel® Nervana™ enables easy experimentation with state of the art Reinforcement Learning algorithms
• homepage: http://coach.nervanasys.com/
• github: https://github.com/NervanaSystems/coach

Blogs

Reinforcement learning’s foundational flaw

https://thegradient.pub/why-rl-is-flawed/

A Short Introduction To Some Reinforcement Learning Algorithms

http://webdocs.cs.ualberta.ca/~vanhasse/rl_algs/rl_algs.html

A Painless Q-Learning Tutorial

http://mnemstudio.org/path-finding-q-learning-tutorial.htm

---

Reinforcement Learning - Part 1

http://outlace.com/Reinforcement-Learning-Part-1/

Reinforcement Learning - Monte Carlo Methods

http://outlace.com/Reinforcement-Learning-Part-2/

Q-learning with Neural Networks

http://outlace.com/Reinforcement-Learning-Part-3/

---

Guest Post (Part I): Demystifying Deep Reinforcement Learning

http://www.nervanasys.com/demystifying-deep-reinforcement-learning/

Using reinforcement learning in Python to teach a virtual car to avoid obstacles: An experiment in Q-learning, neural networks and Pygame.

• blog: https://medium.com/@harvitronix/using-reinforcement-learning-in-python-to-teach-a-virtual-car-to-avoid-obstacles-6e782cc7d4c6#.p8ug6snri
• github: https://github.com/harvitronix/reinforcement-learning-car

Reinforcement learning in Python to teach a virtual car to avoid obstacles — part 2

https://medium.com/@harvitronix/reinforcement-learning-in-python-to-teach-a-virtual-car-to-avoid-obstacles-part-2-93e614fcd238#.i0o643m1h

Some Reinforcement Learning Algorithms in Python, C++

• pan: http://pan.baidu.com/s/1mhcYf3M#path=%252FImplementations%2520of%2520Some%2520Reinforcement%2520Learning%2520Algorithms

learning to do laps with reinforcement learning and neural nets

 

• blog: http://matpalm.com/blog/drivebot/
• github: https://github.com/matpalm/drivebot

Get a taste of reinforcement learning — implement a tic tac toe agent

https://medium.com/@shiyan/get-a-taste-of-reinforcement-learning-implement-a-tic-tac-toe-agent-deda5617b2e4#.59bx71a2h

Best reinforcement learning libraries?

• reddit: https://www.reddit.com/r/MachineLearning/comments/4b2ugc/best_reinforcement_learning_libraries/

Super Simple Reinforcement Learning Tutorial

• part 1: https://medium.com/@awjuliani/super-simple-reinforcement-learning-tutorial-part-1-fd544fab149
• part 2: https://medium.com/@awjuliani/super-simple-reinforcement-learning-tutorial-part-2-ded33892c724#.dyhxww1u6
• part 3: https://medium.com/@awjuliani/simple-reinforcement-learning-with-tensorflow-part-3-model-based-rl-9a6fe0cce99#.r4c7i7tjq
• gist: https://gist.github.com/awjuliani/16608e1c4968baaa692b9b8c7dd94d04

Reinforcement Learning in Python

• github: https://github.com/NathanEpstein/pydata-reinforce

The Skynet Salesman

• keyworkds: traveling salesman problem (TSP), deep Q learning
• blog: http://multithreaded.stitchfix.com/blog/2016/07/21/skynet-salesman/
• github: https://github.com/jn2clark/ReinforcementLearning/tree/master/DeepQ

Apprenticeship learning using Inverse Reinforcement Learning

• blog: https://jangirrishabh.github.io/2016/07/09/virtual-car-IRL/
• github: https://github.com/jangirrishabh/toyCarIRL

Reinforcement Learning and DQN, learning to play from pixels

• blog: https://rubenfiszel.github.io/posts/rl4j/2016-08-24-Reinforcement-Learning-and-DQN.html

Deep Learning in a Nutshell: Reinforcement Learning

https://devblogs.nvidia.com/parallelforall/deep-learning-nutshell-reinforcement-learning/

Write an AI to win at Pong from scratch with Reinforcement Learning

https://medium.com/@dhruvp/how-to-write-a-neural-network-to-play-pong-from-scratch-956b57d4f6e0#.n1pgn9chr

Learning Reinforcement Learning (with Code, Exercises and Solutions)

• blog: http://www.wildml.com/2016/10/learning-reinforcement-learning/
• github: https://github.com/dennybritz/reinforcement-learning

Deep Reinforcement Learning: Playing a Racing Game

https://lopespm.github.io/machine_learning/2016/10/06/deep-reinforcement-learning-racing-game.html

Experimenting with Reinforcement Learning and Active Inference

• blog: http://www.araya.org/archives/955
• github: https://github.com/arayabrain/BinarySearchLSTM

Deep reinforcement learning, battleship

• blog: http://efavdb.com/battleship/
• github: https://github.com/EFavDB/battleship

Deep Learning Research Review Week 2: Reinforcement Learning

https://adeshpande3.github.io/adeshpande3.github.io/Deep-Learning-Research-Review-Week-2-Reinforcement-Learning

Reinforcement Learning: Artificial Intelligence in Game Playing

https://medium.com/@pavelkordik/reinforcement-learning-the-hardest-part-of-machine-learning-b667a22995ca#.jjiitflok

Artificial Intelligence’s Next Big Step: Reinforcement Learning

http://thenewstack.io/reinforcement-learning-ready-real-world/

Let’s make a DQN

Let’s make a DQN

• Theory: https://jaromiru.com/2016/09/27/lets-make-a-dqn-theory/
• Implementation: https://jaromiru.com/2016/10/03/lets-make-a-dqn-implementation/
• Debugging: https://jaromiru.com/2016/10/12/lets-make-a-dqn-debugging/
• Full DQN: https://jaromiru.com/2016/10/21/lets-make-a-dqn-full-dqn/
• github: https://github.com/jaara/AI-blog/blob/master/CartPole-basic.py

Books

Reinforcement Learning: State-of-the-Art

• intro: “The main goal of this book is to present an up-to-date series of survey articles on the main contemporary sub-fields of reinforcement learning. This includes surveys on partially observable environments, hierarchical task decompositions, relational knowledge representation and predictive state representations. Furthermore, topics such as transfer, evolutionary methods and continuous spaces in reinforcement learning are surveyed. In addition, several chapters review reinforcement learning methods in robotics, in games, and in computational neuroscience. In total seventeen different subfields are presented by mostly young experts in those areas, and together they truly represent a state-of-the-art of current reinforcement learning research.”
• book: http://www.springer.com/gp/book/9783642276446#

Reinforcement Learning: An Introduction

• github: https://github.com/Mononofu/reinforcement-learning
• homepage: http://webdocs.cs.ualberta.ca/~sutton/book/the-book.html
• course: http://incompleteideas.net/rlai.cs.ualberta.ca/RLAI/RLAIcourse/2010.html
• book(1st edition): http://pan.baidu.com/s/1jkaMq
• book(2rd edition): http://pan.baidu.com/s/1dDnNEnR

Reinforcement Learning: An Introduction (Second edition, Draft)

• book: https://webdocs.cs.ualberta.ca/~sutton/book/bookdraft2016sep.pdf
• mirror: https://pan.baidu.com/s/1slrMYkP
• github: https://github.com/ShangtongZhang/reinforcement-learning-an-introduction

The Self Learning Quant

• intro: explain and show the concept of self reinforcement learning combined with a neural network
• blog: https://medium.com/@danielzakrisson/the-self-learning-quant-d3329fcc9915#.9lsa5rh3e
• gihtub: https://github.com/danielzak/sl-quant

Reinforcement Learning: An Introduction

• author: Richard S. Sutton and Andrew G. Barto
• book: https://webdocs.cs.ualberta.ca/~sutton/book/ebook/the-book.html
• solutions: https://github.com/btaba/intro-to-rl

Resources

Deep Reinforcement Learning Papers

https://github.com/junhyukoh/deep-reinforcement-learning-papers

Awesome Reinforcement Learning

• website: http://aikorea.org/awesome-rl/?utm_content=buffer5d0f3&utm_medium=social&utm_source=plus.google.com&utm_campaign=buffer#online-demos
• github: https://github.com/aikorea/awesome-rl

Deep Reinforcement Learning Papers

• github: https://github.com/muupan/deep-reinforcement-learning-papers

Deep Reinforcement Learning 深度增强学习资源

• blog: https://zhuanlan.zhihu.com/p/20885568

deep-reinforcement-learning-networks: A list of deep neural network architectures for reinforcement learning tasks

• github: https://github.com/5vision/deep-reinforcement-learning-networks

Deep Reinforcement Learning survey

• github: https://github.com/andrewliao11/Deep-Reinforcement-Learning-Survey

Studying Reinforcement Learning Guide

• github: https://github.com/0bserver07/Study-Reinforcement-Learning

Reading and Questions

What are the best books about reinforcement learning?

https://www.quora.com/What-are-the-best-books-about-reinforcement-learning