awesome-rl
Awesome RL: Papers, Books, Codes, Benchmarks
https://github.com/dbobrenko/awesome-rl
Last synced: 24 days ago
JSON representation
![auto-ml] Auto ML <a name="auto-ml"></a>
Books
![evolution] Evolutionary Algorithms <a name="evolution"></a>
![exploration] Exploration <a name="exploration"></a>
- [Go-Explore
- [Exploration by Random Network Distillation (RND) - learning-with-prediction-based-rewards/)] [code] 2018 @ OpenAI
- navigation - scale-curiosity/)] 2018 @ OpenAI, Berkeley, Univ. of Edinburgh
- [RUDDER: Return Decomposition for Delayed Rewards - jku/baselines-rudder)] 2018 @ Johannes Kepler Univ. Linz
- [Deep Curiosity Search
- locomotion
- transfer - imagine-and-plan/)] 2017 @ DeepMind
![inverse-rl] Inverse RL <a name="inverse-rl"></a>
![locomotion] Locomotion <a name="locomotion"></a>
- [Emergence of Locomotion Behaviours in Rich Environments - flexible-behaviours-simulated-environments/)] Heess et al., 2017 @ DeepMind
- [Programmable Agents
![manipulation] Manipulation <a name="manipulation"></a>
- generalization - dexterity/)] Andrychowicz et al., 2018 @ OpenAI
- generalization - from-simulation/)] Pinto et al., 2017 @ OpenAI, CMU
- generalization - from-simulation/)] Peng et al., 2017 @ OpenAI, Berkeley
![meta-learning] Meta-Learning
- locomotion - a-hierarchy/)] Frans et al., 2017 @ OpenAI, Berkeley.
- [Hybrid Reward Architecture for Reinforcement Learning (HRA)
Misc
![model-based] Model-Based Generic Agents <a name="model-based"></a>
- [Model-Based Reinforcement Learning for Atari
- navigation
- locomotion - based-rl/)] [code] 2017 @ Berkeley
- locomotion - imagine-and-plan/)] 2017 @ Google DeepMind
- navigation
![multi-agent-rl] Multi-Agent RL
- [Learning with Opponent-Learning Awareness (LOLA) - to-model-other-minds/)] Foerster et al., 2017 @ OpenAI, Oxford, Berkeley, CMU
![navigation] Navigation <a name="navigation"></a>
- [Learning to Navigate in Cities Without a Map
- [Human-level performance in first-person multiplayer games with population-based deep reinforcement learning - the-flag/)] Jaderberg et al, 2018 @ DeepMind
- generalization
- [Learning to Navigate in Complex Environments
- transfer
- meta-learning
- [Learning to act by predicting the future (VizDoom 2016 Full DM Winner)
- [Playing FPS Games with Deep Reinforcement Learning (VizDoom 2016 Limited DM 2nd place)
Other Domains <a name="other-domains"></a>
Policy-Based Generic Agents <a name="policy-agents"></a>
- [Soft Actor Critic - actor-critic-deep-reinforcement.html)] [code] 2018 @ Google Brain, UC Berkeley
- [IMPALA
- [Scalable trust-region method for deep reinforcement learning using Kronecker-factored approximation (ACKTR, A2C)
- [Proximal Policy Optimization Algorithms (PPO) - baselines-ppo/)] 2017 @ OpenAI
- [High-dimensional continuous control using generalized advantage estimation (GAE)
- [Trust Region Policy Optimization (TRPO)
- [Actor-Critic Algorithms, pdf
- [Simple Statistical Gradient-Following Algorithms for Connectionist Reinforcement Learning (REINFORCE), pdf
RL Benchmarks <a name="benchmarks"></a>
RL Frameworks & Implementations <a name="frameworks"></a>
Search for new Papers
![self-play] Self-Play <a name="self-play"></a>
Value-Based Generic Agents <a name="value-agents"></a>
- [Implicit Quantile Networks for Distributional Reinforcement Learning (IQN)
- [A Distributional Perspective on Reinforcement Learning (c51)
- [Rainbow: Combining Improvements in Deep Reinforcement Learning
- [Dueling Network Architectures for Deep Reinforcement Learning (Dueling DQN)
- [Deep Reinforcement Learning with Double Q-learning (Double DQN)
- [Playing Atari with Deep Reinforcement Learning** (DQN)
- [Temporal Difference Learning and TD-Gammon, pdf