Deep Reinforcement Learning Papers

A list of recent papers regarding deep reinforcement learning.
The papers are organized based on manually-defined bookmarks.
They are sorted by time to see the recent papers first.
Any suggestions and pull requests are welcome.

Bookmarks

• All Papers
• Value Function Approximation
• Policy Gradient
• Discrete Control
• Continuous Control
• Text Domain
• Visual Domain
• Robotics
• Games
• Monte-Carlo Tree Search
• Inverse Reinforcement Learning
• Improving Exploration
• Transfer Learning
• Multi-Agent

All Papers

• Continuous Deep Q-Learning with Model-based Acceleration, Shixiang Gu et al., arXiv, 2016.
• Deep Exploration via Bootstrapped DQN, I. Osband et al., arXiv, 2016.
• Value Iteration Networks, A. Tamar et al., arXiv, 2016.
• Learning to Communicate to Solve Riddles with Deep Distributed Recurrent Q-Networks, J. N. Foerster et al., arXiv, 2016.
• Asynchronous Methods for Deep Reinforcement Learning, V. Mnih et al., arXiv, 2016.
• Mastering the game of Go with deep neural networks and tree search, D. Silver et al., Nature, 2016.
• Increasing the Action Gap: New Operators for Reinforcement Learning, M. G. Bellemare et al., AAAI, 2016.
• Memory-based control with recurrent neural networks, N. Heess et al., NIPS Workshop, 2015.
• How to Discount Deep Reinforcement Learning: Towards New Dynamic Strategies, V. François-Lavet et al., NIPS Workshop, 2015.
• Multiagent Cooperation and Competition with Deep Reinforcement Learning, A. Tampuu et al., arXiv, 2015.
• Strategic Dialogue Management via Deep Reinforcement Learning, H. Cuayáhuitl et al., NIPS Workshop, 2015.
• MazeBase: A Sandbox for Learning from Games, S. Sukhbaatar et al., arXiv, 2016.
• Learning Simple Algorithms from Examples, W. Zaremba et al., arXiv, 2015.
• Dueling Network Architectures for Deep Reinforcement Learning, Z. Wang et al., arXiv, 2015.
• Actor-Mimic: Deep Multitask and Transfer Reinforcement Learning, E. Parisotto, et al., ICLR, 2016.
• Better Computer Go Player with Neural Network and Long-term Prediction, Y. Tian et al., ICLR, 2016.
• Policy Distillation, A. A. Rusu et at., ICLR, 2016.
• Prioritized Experience Replay, T. Schaul et al., ICLR, 2016.
• Deep Reinforcement Learning with an Action Space Defined by Natural Language, J. He et al., arXiv, 2015.
• Deep Reinforcement Learning in Parameterized Action Space, M. Hausknecht et al., ICLR, 2016.
• Towards Vision-Based Deep Reinforcement Learning for Robotic Motion Control, F. Zhang et al., arXiv, 2015.
• Generating Text with Deep Reinforcement Learning, H. Guo, arXiv, 2015.
• ADAAPT: A Deep Architecture for Adaptive Policy Transfer from Multiple Sources, J. Rajendran et al., arXiv, 2015.
• Variational Information Maximisation for Intrinsically Motivated Reinforcement Learning, S. Mohamed and D. J. Rezende, arXiv, 2015.
• Deep Reinforcement Learning with Double Q-learning, H. van Hasselt et al., arXiv, 2015.
• Recurrent Reinforcement Learning: A Hybrid Approach, X. Li et al., arXiv, 2015.
• Continuous control with deep reinforcement learning, T. P. Lillicrap et al., ICLR, 2016.
• Language Understanding for Text-based Games Using Deep Reinforcement Learning, K. Narasimhan et al., EMNLP, 2015.
• Giraffe: Using Deep Reinforcement Learning to Play Chess, M. Lai, arXiv, 2015.
• Action-Conditional Video Prediction using Deep Networks in Atari Games, J. Oh et al., NIPS, 2015.
• Learning Continuous Control Policies by Stochastic Value Gradients, N. Heess et al., NIPS, 2015.
• Learning Deep Neural Network Policies with Continuous Memory States, M. Zhang et al., arXiv, 2015.
• Deep Recurrent Q-Learning for Partially Observable MDPs, M. Hausknecht and P. Stone, arXiv, 2015.
• Listen, Attend, and Walk: Neural Mapping of Navigational Instructions to Action Sequences, H. Mei et al., arXiv, 2015.
• Incentivizing Exploration In Reinforcement Learning With Deep Predictive Models, B. C. Stadie et al., arXiv, 2015.
• Maximum Entropy Deep Inverse Reinforcement Learning, M. Wulfmeier et al., arXiv, 2015.
• High-Dimensional Continuous Control Using Generalized Advantage Estimation, J. Schulman et al., ICLR, 2016.
• End-to-End Training of Deep Visuomotor Policies, S. Levine et al., arXiv, 2015.
• DeepMPC: Learning Deep Latent Features for Model Predictive Control, I. Lenz, et al., RSS, 2015.
• Universal Value Function Approximators, T. Schaul et al., ICML, 2015.
• Deterministic Policy Gradient Algorithms, D. Silver et al., ICML, 2015.
• Massively Parallel Methods for Deep Reinforcement Learning, A. Nair et al., ICML Workshop, 2015.
• Trust Region Policy Optimization, J. Schulman et al., ICML, 2015.
• Human-level control through deep reinforcement learning, V. Mnih et al., Nature, 2015.
• Deep Learning for Real-Time Atari Game Play Using Offline Monte-Carlo Tree Search Planning, X. Guo et al., NIPS, 2014.
• Playing Atari with Deep Reinforcement Learning, V. Mnih et al., NIPS Workshop, 2013.

Value Function Approximation

• Continuous Deep Q-Learning with Model-based Acceleration, Shixiang Gu et al., arXiv, 2016.
• Deep Exploration via Bootstrapped DQN, I. Osband et al., arXiv, 2016.
• Value Iteration Networks, A. Tamar et al., arXiv, 2016.
• Learning to Communicate to Solve Riddles with Deep Distributed Recurrent Q-Networks, J. N. Foerster et al., arXiv, 2016.
• Asynchronous Methods for Deep Reinforcement Learning, V. Mnih et al., arXiv, 2016.
• Mastering the game of Go with deep neural networks and tree search, D. Silver et al., Nature, 2016.
• Increasing the Action Gap: New Operators for Reinforcement Learning, M. G. Bellemare et al., AAAI, 2016.
• How to Discount Deep Reinforcement Learning: Towards New Dynamic Strategies, V. François-Lavet et al., NIPS Workshop, 2015.
• Multiagent Cooperation and Competition with Deep Reinforcement Learning, A. Tampuu et al., arXiv, 2015.
• Strategic Dialogue Management via Deep Reinforcement Learning, H. Cuayáhuitl et al., NIPS Workshop, 2015.
• Learning Simple Algorithms from Examples, W. Zaremba et al., arXiv, 2015.
• Dueling Network Architectures for Deep Reinforcement Learning, Z. Wang et al., arXiv, 2015.
• Prioritized Experience Replay, T. Schaul et al., ICLR, 2016.
• Deep Reinforcement Learning with an Action Space Defined by Natural Language, J. He et al., arXiv, 2015.
• Deep Reinforcement Learning in Parameterized Action Space, M. Hausknecht et al., ICLR, 2016.
• Towards Vision-Based Deep Reinforcement Learning for Robotic Motion Control, F. Zhang et al., arXiv, 2015.
• Generating Text with Deep Reinforcement Learning, H. Guo, arXiv, 2015.
• Deep Reinforcement Learning with Double Q-learning, H. van Hasselt et al., arXiv, 2015.
• Recurrent Reinforcement Learning: A Hybrid Approach, X. Li et al., arXiv, 2015.
• Continuous control with deep reinforcement learning, T. P. Lillicrap et al., ICLR, 2016.
• Language Understanding for Text-based Games Using Deep Reinforcement Learning, K. Narasimhan et al., EMNLP, 2015.
• Action-Conditional Video Prediction using Deep Networks in Atari Games, J. Oh et al., NIPS, 2015.
• Deep Recurrent Q-Learning for Partially Observable MDPs, M. Hausknecht and P. Stone, arXiv, 2015.
• Incentivizing Exploration In Reinforcement Learning With Deep Predictive Models, B. C. Stadie et al., arXiv, 2015.
• Massively Parallel Methods for Deep Reinforcement Learning, A. Nair et al., ICML Workshop, 2015.
• Human-level control through deep reinforcement learning, V. Mnih et al., Nature, 2015.
• Playing Atari with Deep Reinforcement Learning, V. Mnih et al., NIPS Workshop, 2013.

Policy Gradient

• Asynchronous Methods for Deep Reinforcement Learning, V. Mnih et al., arXiv, 2016.
• Mastering the game of Go with deep neural networks and tree search, D. Silver et al., Nature, 2016.
• Memory-based control with recurrent neural networks, N. Heess et al., NIPS Workshop, 2015.
• MazeBase: A Sandbox for Learning from Games, S. Sukhbaatar et al., arXiv, 2016.
• ADAAPT: A Deep Architecture for Adaptive Policy Transfer from Multiple Sources, J. Rajendran et al., arXiv, 2015.
• Continuous control with deep reinforcement learning, T. P. Lillicrap et al., ICLR, 2016.
• Learning Continuous Control Policies by Stochastic Value Gradients, N. Heess et al., NIPS, 2015.
• High-Dimensional Continuous Control Using Generalized Advantage Estimation, J. Schulman et al., ICLR, 2016.
• End-to-End Training of Deep Visuomotor Policies, S. Levine et al., arXiv, 2015.
• Deterministic Policy Gradient Algorithms, D. Silver et al., ICML, 2015.
• Trust Region Policy Optimization, J. Schulman et al., ICML, 2015.

Discrete Control

• Deep Exploration via Bootstrapped DQN, I. Osband et al., arXiv, 2016.
• Value Iteration Networks, A. Tamar et al., arXiv, 2016.
• Learning to Communicate to Solve Riddles with Deep Distributed Recurrent Q-Networks, J. N. Foerster et al., arXiv, 2016.
• Asynchronous Methods for Deep Reinforcement Learning, V. Mnih et al., arXiv, 2016.
• Mastering the game of Go with deep neural networks and tree search, D. Silver et al., Nature, 2016.
• Increasing the Action Gap: New Operators for Reinforcement Learning, M. G. Bellemare et al., AAAI, 2016.
• How to Discount Deep Reinforcement Learning: Towards New Dynamic Strategies, V. François-Lavet et al., NIPS Workshop, 2015.
• Multiagent Cooperation and Competition with Deep Reinforcement Learning, A. Tampuu et al., arXiv, 2015.
• Strategic Dialogue Management via Deep Reinforcement Learning, H. Cuayáhuitl et al., NIPS Workshop, 2015.
• Learning Simple Algorithms from Examples, W. Zaremba et al., arXiv, 2015.
• Dueling Network Architectures for Deep Reinforcement Learning, Z. Wang et al., arXiv, 2015.
• Better Computer Go Player with Neural Network and Long-term Prediction, Y. Tian et al., ICLR, 2016.
• Actor-Mimic: Deep Multitask and Transfer Reinforcement Learning, E. Parisotto, et al., ICLR, 2016.
• Policy Distillation, A. A. Rusu et at., ICLR, 2016.
• Prioritized Experience Replay, T. Schaul et al., ICLR, 2016.
• Deep Reinforcement Learning with an Action Space Defined by Natural Language, J. He et al., arXiv, 2015.
• Deep Reinforcement Learning in Parameterized Action Space, M. Hausknecht et al., ICLR, 2016.
• Towards Vision-Based Deep Reinforcement Learning for Robotic Motion Control, F. Zhang et al., arXiv, 2015.
• Generating Text with Deep Reinforcement Learning, H. Guo, arXiv, 2015.
• ADAAPT: A Deep Architecture for Adaptive Policy Transfer from Multiple Sources, J. Rajendran et al., arXiv, 2015.
• Variational Information Maximisation for Intrinsically Motivated Reinforcement Learning, S. Mohamed and D. J. Rezende, arXiv, 2015.
• Deep Reinforcement Learning with Double Q-learning, H. van Hasselt et al., arXiv, 2015.
• Recurrent Reinforcement Learning: A Hybrid Approach, X. Li et al., arXiv, 2015.
• Language Understanding for Text-based Games Using Deep Reinforcement Learning, K. Narasimhan et al., EMNLP, 2015.
• Giraffe: Using Deep Reinforcement Learning to Play Chess, M. Lai, arXiv, 2015.
• Action-Conditional Video Prediction using Deep Networks in Atari Games, J. Oh et al., NIPS, 2015.
• Deep Recurrent Q-Learning for Partially Observable MDPs, M. Hausknecht and P. Stone, arXiv, 2015.
• Listen, Attend, and Walk: Neural Mapping of Navigational Instructions to Action Sequences, H. Mei et al., arXiv, 2015.
• Incentivizing Exploration In Reinforcement Learning With Deep Predictive Models, B. C. Stadie et al., arXiv, 2015.
• Universal Value Function Approximators, T. Schaul et al., ICML, 2015.
• Massively Parallel Methods for Deep Reinforcement Learning, A. Nair et al., ICML Workshop, 2015.
• Human-level control through deep reinforcement learning, V. Mnih et al., Nature, 2015.
• Deep Learning for Real-Time Atari Game Play Using Offline Monte-Carlo Tree Search Planning, X. Guo et al., NIPS, 2014.
• Playing Atari with Deep Reinforcement Learning, V. Mnih et al., NIPS Workshop, 2013.

Continuous Control

• Continuous Deep Q-Learning with Model-based Acceleration, Shixiang Gu et al., arXiv, 2016.
• Asynchronous Methods for Deep Reinforcement Learning, V. Mnih et al., arXiv, 2016.
• Memory-based control with recurrent neural networks, N. Heess et al., NIPS Workshop, 2015.
• Variational Information Maximisation for Intrinsically Motivated Reinforcement Learning, S. Mohamed and D. J. Rezende, arXiv, 2015.
• Continuous control with deep reinforcement learning, T. P. Lillicrap et al., ICLR, 2016.
• Learning Continuous Control Policies by Stochastic Value Gradients, N. Heess et al., NIPS, 2015.
• Learning Deep Neural Network Policies with Continuous Memory States, M. Zhang et al., arXiv, 2015.
• High-Dimensional Continuous Control Using Generalized Advantage Estimation, J. Schulman et al., ICLR, 2016.
• End-to-End Training of Deep Visuomotor Policies, S. Levine et al., arXiv, 2015.
• DeepMPC: Learning Deep Latent Features for Model Predictive Control, I. Lenz, et al., RSS, 2015.
• Deterministic Policy Gradient Algorithms, D. Silver et al., ICML, 2015.
• Trust Region Policy Optimization, J. Schulman et al., ICML, 2015.

Text Domain

• Strategic Dialogue Management via Deep Reinforcement Learning, H. Cuayáhuitl et al., NIPS Workshop, 2015.
• MazeBase: A Sandbox for Learning from Games, S. Sukhbaatar et al., arXiv, 2016.
• Deep Reinforcement Learning with an Action Space Defined by Natural Language, J. He et al., arXiv, 2015.
• Generating Text with Deep Reinforcement Learning, H. Guo, arXiv, 2015.
• Language Understanding for Text-based Games Using Deep Reinforcement Learning, K. Narasimhan et al., EMNLP, 2015.
• Listen, Attend, and Walk: Neural Mapping of Navigational Instructions to Action Sequences, H. Mei et al., arXiv, 2015.

Visual Domain

• Deep Exploration via Bootstrapped DQN, I. Osband et al., arXiv, 2016.
• Value Iteration Networks, A. Tamar et al., arXiv, 2016.
• Asynchronous Methods for Deep Reinforcement Learning, V. Mnih et al., arXiv, 2016.
• Mastering the game of Go with deep neural networks and tree search, D. Silver et al., Nature, 2016.
• Increasing the Action Gap: New Operators for Reinforcement Learning, M. G. Bellemare et al., AAAI, 2016.
• Memory-based control with recurrent neural networks, N. Heess et al., NIPS Workshop, 2015.
• How to Discount Deep Reinforcement Learning: Towards New Dynamic Strategies, V. François-Lavet et al., NIPS Workshop, 2015.
• Multiagent Cooperation and Competition with Deep Reinforcement Learning, A. Tampuu et al., arXiv, 2015.
• Dueling Network Architectures for Deep Reinforcement Learning, Z. Wang et al., arXiv, 2015.
• Actor-Mimic: Deep Multitask and Transfer Reinforcement Learning, E. Parisotto, et al., ICLR, 2016.
• Better Computer Go Player with Neural Network and Long-term Prediction, Y. Tian et al., ICLR, 2016.
• Policy Distillation, A. A. Rusu et at., ICLR, 2016.
• Prioritized Experience Replay, T. Schaul et al., ICLR, 2016.
• Deep Reinforcement Learning in Parameterized Action Space, M. Hausknecht et al., ICLR, 2016.
• Towards Vision-Based Deep Reinforcement Learning for Robotic Motion Control, F. Zhang et al., arXiv, 2015.
• Variational Information Maximisation for Intrinsically Motivated Reinforcement Learning, S. Mohamed and D. J. Rezende, arXiv, 2015.
• Deep Reinforcement Learning with Double Q-learning, H. van Hasselt et al., arXiv, 2015.
• Continuous control with deep reinforcement learning, T. P. Lillicrap et al., ICLR, 2016.
• Giraffe: Using Deep Reinforcement Learning to Play Chess, M. Lai, arXiv, 2015.
• Action-Conditional Video Prediction using Deep Networks in Atari Games, J. Oh et al., NIPS, 2015.
• Learning Continuous Control Policies by Stochastic Value Gradients, N. Heess et al., NIPS, 2015.
• Deep Recurrent Q-Learning for Partially Observable MDPs, M. Hausknecht and P. Stone, arXiv, 2015.
• Incentivizing Exploration In Reinforcement Learning With Deep Predictive Models, B. C. Stadie et al., arXiv, 2015.
• High-Dimensional Continuous Control Using Generalized Advantage Estimation, J. Schulman et al., ICLR, 2016.
• End-to-End Training of Deep Visuomotor Policies, S. Levine et al., arXiv, 2015.
• Universal Value Function Approximators, T. Schaul et al., ICML, 2015.
• Massively Parallel Methods for Deep Reinforcement Learning, A. Nair et al., ICML Workshop, 2015.
• Trust Region Policy Optimization, J. Schulman et al., ICML, 2015.
• Human-level control through deep reinforcement learning, V. Mnih et al., Nature, 2015.
• Deep Learning for Real-Time Atari Game Play Using Offline Monte-Carlo Tree Search Planning, X. Guo et al., NIPS, 2014.
• Playing Atari with Deep Reinforcement Learning, V. Mnih et al., NIPS Workshop, 2013.

Robotics

• Continuous Deep Q-Learning with Model-based Acceleration, Shixiang Gu et al., arXiv, 2016.
• Asynchronous Methods for Deep Reinforcement Learning, V. Mnih et al., arXiv, 2016.
• Memory-based control with recurrent neural networks, N. Heess et al., NIPS Workshop, 2015.
• Towards Vision-Based Deep Reinforcement Learning for Robotic Motion Control, F. Zhang et al., arXiv, 2015.
• Learning Continuous Control Policies by Stochastic Value Gradients, N. Heess et al., NIPS, 2015.
• Learning Deep Neural Network Policies with Continuous Memory States, M. Zhang et al., arXiv, 2015.
• High-Dimensional Continuous Control Using Generalized Advantage Estimation, J. Schulman et al., ICLR, 2016.
• End-to-End Training of Deep Visuomotor Policies, S. Levine et al., arXiv, 2015.
• DeepMPC: Learning Deep Latent Features for Model Predictive Control, I. Lenz, et al., RSS, 2015.
• Trust Region Policy Optimization, J. Schulman et al., ICML, 2015.

Games

• Deep Exploration via Bootstrapped DQN, I. Osband et al., arXiv, 2016.
• Learning to Communicate to Solve Riddles with Deep Distributed Recurrent Q-Networks, J. N. Foerster et al., arXiv, 2016.
• Asynchronous Methods for Deep Reinforcement Learning, V. Mnih et al., arXiv, 2016.
• Mastering the game of Go with deep neural networks and tree search, D. Silver et al., Nature, 2016.
• Increasing the Action Gap: New Operators for Reinforcement Learning, M. G. Bellemare et al., AAAI, 2016.
• How to Discount Deep Reinforcement Learning: Towards New Dynamic Strategies, V. François-Lavet et al., NIPS Workshop, 2015.
• Multiagent Cooperation and Competition with Deep Reinforcement Learning, A. Tampuu et al., arXiv, 2015.
• MazeBase: A Sandbox for Learning from Games, S. Sukhbaatar et al., arXiv, 2016.
• Dueling Network Architectures for Deep Reinforcement Learning, Z. Wang et al., arXiv, 2015.
• Better Computer Go Player with Neural Network and Long-term Prediction, Y. Tian et al., ICLR, 2016.
• Actor-Mimic: Deep Multitask and Transfer Reinforcement Learning, E. Parisotto, et al., ICLR, 2016.
• Policy Distillation, A. A. Rusu et at., ICLR, 2016.
• Prioritized Experience Replay, T. Schaul et al., ICLR, 2016.
• Deep Reinforcement Learning with an Action Space Defined by Natural Language, J. He et al., arXiv, 2015.
• Deep Reinforcement Learning in Parameterized Action Space, M. Hausknecht et al., ICLR, 2016.
• Variational Information Maximisation for Intrinsically Motivated Reinforcement Learning, S. Mohamed and D. J. Rezende, arXiv, 2015.
• Deep Reinforcement Learning with Double Q-learning, H. van Hasselt et al., arXiv, 2015.
• Continuous control with deep reinforcement learning, T. P. Lillicrap et al., ICLR, 2016.
• Language Understanding for Text-based Games Using Deep Reinforcement Learning, K. Narasimhan et al., EMNLP, 2015.
• Giraffe: Using Deep Reinforcement Learning to Play Chess, M. Lai, arXiv, 2015.
• Action-Conditional Video Prediction using Deep Networks in Atari Games, J. Oh et al., NIPS, 2015.
• Deep Recurrent Q-Learning for Partially Observable MDPs, M. Hausknecht and P. Stone, arXiv, 2015.
• Incentivizing Exploration In Reinforcement Learning With Deep Predictive Models, B. C. Stadie et al., arXiv, 2015.
• Universal Value Function Approximators, T. Schaul et al., ICML, 2015.
• Massively Parallel Methods for Deep Reinforcement Learning, A. Nair et al., ICML Workshop, 2015.
• Trust Region Policy Optimization, J. Schulman et al., ICML, 2015.
• Human-level control through deep reinforcement learning, V. Mnih et al., Nature, 2015.
• Deep Learning for Real-Time Atari Game Play Using Offline Monte-Carlo Tree Search Planning, X. Guo et al., NIPS, 2014.
• Playing Atari with Deep Reinforcement Learning, V. Mnih et al., NIPS Workshop, 2013.

Monte-Carlo Tree Search

• Mastering the game of Go with deep neural networks and tree search, D. Silver et al., Nature, 2016.
• Better Computer Go Player with Neural Network and Long-term Prediction, Y. Tian et al., ICLR, 2016.
• Deep Learning for Real-Time Atari Game Play Using Offline Monte-Carlo Tree Search Planning, X. Guo et al., NIPS, 2014.

Inverse Reinforcement Learning

• Maximum Entropy Deep Inverse Reinforcement Learning, M. Wulfmeier et al., arXiv, 2015.

Transfer Learning

• Actor-Mimic: Deep Multitask and Transfer Reinforcement Learning, E. Parisotto, et al., ICLR, 2016.
• Policy Distillation, A. A. Rusu et at., ICLR, 2016.
• ADAAPT: A Deep Architecture for Adaptive Policy Transfer from Multiple Sources, J. Rajendran et al., arXiv, 2015.
• Universal Value Function Approximators, T. Schaul et al., ICML, 2015.

Improving Exploration

• Deep Exploration via Bootstrapped DQN, I. Osband et al., arXiv, 2016.
• Action-Conditional Video Prediction using Deep Networks in Atari Games, J. Oh et al., NIPS, 2015.
• Incentivizing Exploration In Reinforcement Learning With Deep Predictive Models, B. C. Stadie et al., arXiv, 2015.

Multi-Agent

• Learning to Communicate to Solve Riddles with Deep Distributed Recurrent Q-Networks, J. N. Foerster et al., arXiv, 2016.
• Multiagent Cooperation and Competition with Deep Reinforcement Learning, A. Tampuu et al., arXiv, 2015.