└── README.md /README.md: -------------------------------------------------------------------------------- 1 | # Robust-RL-Baselines 2 | 3 | 4 | 5 | #### 1. Robust Single Agent RL Baselines 6 | - Breaking the Barrier: Enhanced Utility and Robustness in Smoothed DRL Agents, [Paper](https://arxiv.org/pdf/2406.18062), [Code](https://github.com/Trustworthy-ML-Lab/Robust_HighUtil_Smoothed_DRL) 7 | - Fine-Grained Causal Dynamics Learning with Quantization for Improving Robustness in Reinforcement Learning, [Paper](https://arxiv.org/pdf/2406.03234), [Code](https://github.com/iwhwang/Fine-Grained-Causal-RL), (Accepted by ICML 2024) 8 | - Towards Robust Offline Reinforcement Learning under Diverse Data Corruption, [Paper](https://arxiv.org/pdf/2310.12955), [Code](https://github.com/YangRui2015/RIQL), (Accepted by ICLR 2024) 9 | - Robust Offline Reinforcement Learning with Heavy-Tailed Rewards, [Paper](https://proceedings.mlr.press/v238/zhu24a/zhu24a.pdf), [Code](https://github.com/Mamba413/ROOM), (Accepted by AISTATS 2024) 10 | - Causal Counterfactuals for Improving the Robustness of Reinforcement Learning, [Paper](https://arxiv.org/pdf/2211.05551), [Code](https://github.com/Tom1042roboai/CausalCF), (Accepted by AAMAS 2023) 11 | - RAMRL: Towards Robust On-Ramp Merging via Augmented Multimodal Reinforcement Learning, [Paper](https://arxiv.org/pdf/2208.07307), [Code](https://github.com/grbagwe/RAMRL), (Accepted by MOST 2023) 12 | - Robust Deep Reinforcement Learning through Bootstrapped Opportunistic Curriculum, [Paper](https://arxiv.org/pdf/2206.10057.pdf), [Code](https://github.com/jlwu002/BCL), (Accepted by ICML 2022) 13 | - Robust offline Reinforcement Learning via Conservative Smoothing, [Paper](https://arxiv.org/pdf/2206.02829), [Code](https://github.com/YangRui2015/RORL), (Accepted by NeurIPS 2022) 14 | - Robust On-Policy Sampling for Data-Efficient Policy Evaluation in Reinforcement Learning, [Paper](https://arxiv.org/pdf/2111.14552), [Code](https://github.com/uoe-agents/robust_onpolicy_data_collection), (Accepted by NeurIPS 2022) 15 | - Robust Reinforcement Learning using Offline Data, [Paper](https://arxiv.org/pdf/2208.05129), [Code](https://github.com/zaiyan-x/RFQI), (Accepted by NeurIPS 2022) 16 | - CROP: Certifying Robust Policies for Reinforcement Learning through Functional Smoothing, [Paper](https://arxiv.org/pdf/2106.09292), [Code](https://github.com/AI-secure/CROP), (Accepted by ICLR 2022) 17 | - COPA: Certifying Robust Policies for Offline Reinforcement Learning against Poisoning Attacks, [Paper](https://arxiv.org/pdf/2203.08398), [Code](https://github.com/AI-secure/COPA), (Accepted by ICLR 2022) 18 | - Robust Risk-Aware Reinforcement Learning, [Paper](https://arxiv.org/pdf/2108.10403), [Code](https://github.com/sebjai/robust-risk-aware-rl), (Accepted by SIAM Journal on Financial Mathematics, 2022) 19 | - Efficient Adversarial Training without Attacking: Worst-Case-Aware Robust Reinforcement Learning, [Paper](https://arxiv.org/pdf/2210.05927), [Code](https://github.com/umd-huang-lab/WocaR-RL), (Accepted by NeurIPS 2022) 20 | - Robust Deep Reinforcement Learning through Adversarial Loss, [Paper](https://arxiv.org/pdf/2008.01976), [Code](https://github.com/tuomaso/radial_rl_v2), (Accepted by NeurIPS 2021) 21 | - Robust Reinforcement Learning with Alternating Training of Learned Adversaries, [Paper](https://arxiv.org/pdf/2101.08452), [Code](https://github.com/huanzhang12/ATLA_robust_RL), (Accepted by ICLR 2021) 22 | - Multi-Modal Mutual Information (MuMMI) Training for Robust Self-Supervised Deep Reinforcement Learning, [Paper](https://arxiv.org/pdf/2107.02339), [Code](https://github.com/clear-nus/MuMMI), (Accepted by ICRA 2021) 23 | - Robust Reinforcement Learning Under Minimax Regret for Green Security, [Paper](https://arxiv.org/pdf/2106.08413), [Code](https://github.com/lily-x/mirror), (Accepted by UAI 2021) 24 | - Robust Reinforcement Learning via Adversarial training with Langevin Dynamics, [Paper](https://arxiv.org/pdf/2002.06063), [Code](https://github.com/ythuangyt/Robust-Reinforcement-Learning-via-Adversarial-training-with-Langevin-Dynamics), (Accepted by NeurIPS 2020) 25 | - Robust Deep Reinforcement Learning against Adversarial Perturbations on State Observations, [Paper](https://arxiv.org/pdf/2003.08938), [Code](https://github.com/chenhongge/StateAdvDRL), (Accepted by NeurIPS 2020) 26 | - Action Robust Reinforcement Learning and Applications in Continuous Control, [Paper](https://arxiv.org/pdf/1901.09184), [Code](https://github.com/tesslerc/ActionRobustRL), (Accepted by ICML 2019) 27 | - Robust Domain Randomization for Reinforcement Learning, [Paper](https://arxiv.org/pdf/1910.10537), [Code](https://github.com/uncharted-technologies/robust-domain-randomization), (Arxiv, 2019) 28 | - Robust Adversarial Reinforcement Learning, [Paper](https://arxiv.org/pdf/1703.02702), [Code](https://github.com/jerinphilip/robust-adversarial-rl), (Accepted by ICML 2015) 29 | 30 | 31 | 32 | #### 2. Robust Multi-Agent RL Baselines 33 | 34 | - Robust Multi-Agent Reinforcement Learning with State Uncertainty, [Paper](https://openreview.net/pdf?id=CqTkapZ6H9), [Code](https://github.com/SihongHo/Robust_MARL_with_State_Uncertainty), (Accepted by TMLR 2023) 35 | - Weaponizing Actions in Multi-Agent Reinforcement Learning: Theoretical and Empirical Study on Security and Robustness [Paper](https://link.springer.com/chapter/10.1007/978-3-031-21203-1_21), [Code](https://github.com/frank47ltt/MARL_Robustness), (Accepted by PRIMA 2022) 36 | --------------------------------------------------------------------------------