Robust-RL-Baselines

1. Robust Single Agent RL Baselines

• Breaking the Barrier: Enhanced Utility and Robustness in Smoothed DRL Agents, Paper, Code
• Fine-Grained Causal Dynamics Learning with Quantization for Improving Robustness in Reinforcement Learning, Paper, Code, (Accepted by ICML 2024)
• Towards Robust Offline Reinforcement Learning under Diverse Data Corruption, Paper, Code, (Accepted by ICLR 2024)
• Robust Offline Reinforcement Learning with Heavy-Tailed Rewards, Paper, Code, (Accepted by AISTATS 2024)
• Causal Counterfactuals for Improving the Robustness of Reinforcement Learning, Paper, Code, (Accepted by AAMAS 2023)
• RAMRL: Towards Robust On-Ramp Merging via Augmented Multimodal Reinforcement Learning, Paper, Code, (Accepted by MOST 2023)
• Robust Deep Reinforcement Learning through Bootstrapped Opportunistic Curriculum, Paper, Code, (Accepted by ICML 2022)
• Robust offline Reinforcement Learning via Conservative Smoothing, Paper, Code, (Accepted by NeurIPS 2022)
• Robust On-Policy Sampling for Data-Efficient Policy Evaluation in Reinforcement Learning, Paper, Code, (Accepted by NeurIPS 2022)
• Robust Reinforcement Learning using Offline Data, Paper, Code, (Accepted by NeurIPS 2022)
• CROP: Certifying Robust Policies for Reinforcement Learning through Functional Smoothing, Paper, Code, (Accepted by ICLR 2022)
• COPA: Certifying Robust Policies for Offline Reinforcement Learning against Poisoning Attacks, Paper, Code, (Accepted by ICLR 2022)
• Robust Risk-Aware Reinforcement Learning, Paper, Code, (Accepted by SIAM Journal on Financial Mathematics, 2022)
• Efficient Adversarial Training without Attacking: Worst-Case-Aware Robust Reinforcement Learning, Paper, Code, (Accepted by NeurIPS 2022)
• Robust Deep Reinforcement Learning through Adversarial Loss, Paper, Code, (Accepted by NeurIPS 2021)
• Robust Reinforcement Learning with Alternating Training of Learned Adversaries, Paper, Code, (Accepted by ICLR 2021)
• Multi-Modal Mutual Information (MuMMI) Training for Robust Self-Supervised Deep Reinforcement Learning, Paper, Code, (Accepted by ICRA 2021)
• Robust Reinforcement Learning Under Minimax Regret for Green Security, Paper, Code, (Accepted by UAI 2021)
• Robust Reinforcement Learning via Adversarial training with Langevin Dynamics, Paper, Code, (Accepted by NeurIPS 2020)
• Robust Deep Reinforcement Learning against Adversarial Perturbations on State Observations, Paper, Code, (Accepted by NeurIPS 2020)
• Action Robust Reinforcement Learning and Applications in Continuous Control, Paper, Code, (Accepted by ICML 2019)
• Robust Domain Randomization for Reinforcement Learning, Paper, Code, (Arxiv, 2019)
• Robust Adversarial Reinforcement Learning, Paper, Code, (Accepted by ICML 2015)

2. Robust Multi-Agent RL Baselines

• Robust Multi-Agent Reinforcement Learning with State Uncertainty, Paper, Code, (Accepted by TMLR 2023)
• Weaponizing Actions in Multi-Agent Reinforcement Learning: Theoretical and Empirical Study on Security and Robustness Paper, Code, (Accepted by PRIMA 2022)