Out of the Box

이 문서는 읽기 전용입니다. 원본을 볼 수는 있지만 바꿀 수는 없습니다. 문제가 있다고 생각하면 관리자에게 문의하세요.
===== General Video Game Playing =====

==== Learning Algorithm ====

  * [[https://arxiv.org/abs/1704.04651|The Reactor: A Sample-Efficient Actor-Critic Architecture]]

==== Reward Shaping ====

  * [[https://arxiv.org/abs/1611.05397|Reinforcement Learning with Unsupervised Auxiliary Tasks]]

==== Simulation ====

  * [[https://arxiv.org/abs/1704.02254|RECURRENT ENVIRONMENT SIMULATORS]]
  * [[https://arxiv.org/abs/1507.08750|Action-Conditional Video Prediction using Deep Networks in Atari Games]]
  * 

==== Learning by Instruction ====

  * [[https://arxiv.org/abs/1704.05539|Beating Atari with Natural Language Guided Reinforcement Learning]]
==== ?? =====

  * ??
    * Z. C. Lipton, J. Gao, L. Li, X. Li, F. Ahmed, and L. Deng, “Efficient Exploration for Dialogue Policy Learning with BBQ Networks & Replay Buffer Spiking,” arXiv:1608.05081 [cs, stat], Aug. 2016.
  * Adaptive Normalization
    * H. van Hasselt et al., “Learning values across many orders of magnitude,” arXiv:1602.07714 [cs, stat], Feb. 2016.
  * [[deep learning:safe off policy|Safe Off-Policy]]
    *R. Munos et al., “Safe and Efficient Off-Policy Reinforcement Learning,” arXiv preprint arXiv:1606.02647, 2016.
  * Empowerment
    * S. Mohamed and D. J. Rezende, “Variational information maximisation for intrinsically motivated reinforcement learning,” Advances in Neural Information Processing Systems, pp. 2125–2133, 2015.
  * Universal Values
    * T. Schaul et al.,  “Universal value function approximators,” in Proceedings of the 32nd International Conference on Machine Learning (ICML-15), 2015, pp. 1312–1320.
  * Macro-Actions
    * A. Vezhnevets et al., “Strategic Attentive Writer for Learning Macro-Actions,” in Advances in Neural Information Processing Systems 29, D. D. Lee, M. Sugiyama, U. V. Luxburg, I. Guyon, and R. Garnett, Eds. Curran Associates, Inc., 2016, pp. 3486–3494.
  * Successor Features
    * A. Barreto et al., “Successor Features for Transfer in Reinforcement Learning,” arXiv preprint arXiv:1606.05312, 2016.
  * Progressive Network
    * A. A. Rusu et al., “Progressive neural networks,” arXiv preprint arXiv:1606.04671, 2016.
  * [[https://arxiv.org/pdf/1310.8499.pdf|2014-05, Deep AutoRegressive Networks]]


==== Deep Reinforcement Learning 알고리즘 ====

^Deep Q-Learning |V. Mnih et al., “Human-level control through deep reinforcement learning,” Nature, vol. 518, no. 7540, pp. 529–533, Feb. 2015. |
^Double DQN |H. van Hasselt, A. Guez, and D. Silver, “Deep Reinforcement Learning with Double Q-learning,” arXiv:1509.06461 [cs], Sep. 2015. |
^DDPG|T. P. Lillicrap et al., “Continuous control with deep reinforcement learning,” arXiv:1509.02971 [cs, stat], Sep. 2015.| 
^Async. Deep RL| V. Mnih et al., “Asynchronous Methods for Deep Reinforcement Learning,” arXiv:1602.01783 [cs], Feb. 2016.|

==== Exploration 개선 ====

^Count & Exploration |I. Osband, C. Blundell, A. Pritzel, and B. Van Roy, “Deep Exploration via Bootstrapped DQN,” arXiv:1602.04621 [cs, stat], Feb. 2016. |
^ | |

==== Replay memory 개선 ====

^Prioritized Replay |T. Schaul, J. Quan, I. Antonoglou, and D. Silver, “Prioritized Experience Replay,” arXiv:1511.05952 [cs], Nov. 2015.|
^ | |

==== Architecture ====

^Dueling Network | Z. Wang, N. de Freitas, and M. Lanctot, “Dueling network architectures for deep reinforcement learning,” arXiv preprint arXiv:1511.06581, 2015.|




==== AI Platform ====

^ViZDoom |M. Kempka, M. Wydmuch, G. Runc, J. Toczek, and W. Jaśkowski, “ViZDoom: A Doom-based AI Research Platform for Visual Reinforcement Learning,” arXiv:1605.02097 [cs], May 2016. |
^OpenAIGym|https://gym.openai.com/ |
^Universe|https://universe.openai.com/|
^DeepMind Lab|https://deepmind.com/blog/open-sourcing-deepmind-lab/|
^Malmo|https://github.com/Microsoft/malmo|

==== Application ====

^AlphaGo|D. Silver et al., “Mastering the game of Go with deep neural networks and tree search,” Nature, vol. 529, no. 7587, pp. 484–489, Jan. 2016.|
^Anold |G. Lample and D. S. Chaplot, “Playing FPS Games with Deep Reinforcement Learning,” arXiv:1609.05521 [cs], Sep. 2016. |

===== 구현체 =====

  * https://github.com/google/dopamine