Out of the Box

사용자 도구

사이트 도구

추적:
drl:start

차이

문서의 선택한 두 판 사이의 차이를 보여줍니다.

차이 보기로 링크

양쪽 이전 판이전 판
다음 판		이전 판
drl:start [2016/11/30 06:23] – 제거됨 rex8312drl:start [2024/03/23 02:42] (현재) – 바깥 편집 127.0.0.1
줄 1: 줄 1:
 +===== 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
 +  
 +
 +