SafeRL-Lab / Trusted-RL-for-Autonomous-Driving

Trusted reinforcement learning algorithms for autonomous driving.

Trusted-RL-for-Autonomous-Driving

The repository is for Trusted Reinforcement Learning (RL) research and its application in autonomous driving, in which we investigate various trusted RL baselines and safe RL benchmarks, including single agent RL and multi-agent RL. If any authors do not want their paper to be listed here, please feel free to contact <gshangd[AT]foxmail.com>. (This repository is under actively development. We appreciate any constructive comments and suggestions)

You are more than welcome to update this list! If you find a paper about Safe RL which is not listed here, please

• fork this repository, add it and merge back;
• or report an issue here;
• or email <gshangd[AT]foxmail.com>.

Uncertainty in RL

• Robust Multi-Agent Reinforcement Learning with Model Uncertainty, Paper, Not Find Code (Accepted by NeurIPS 2020)

• Towards Safe Reinforcement Learning via Constraining Conditional Value-at-Risk, Paper, Not Find Code (Arxiv, 2022)

• Safe, Multi-Agent, Reinforcement Learning for Autonomous Driving, Paper, Not Find Code (Arxiv, Citation 300+, 2016)

• Motion Planning for Autonomous Vehicles in the Presence of Uncertainty Using Reinforcement Learning, Paper, Not Find Code (Accepted by IEEE International Workshop on Intelligent Robots and Systems, 2021)

• Reachability-based Trajectory Safeguard (RTS): A Safe and Fast Reinforcement Learning Safety Layer for Continuous Control, Paper, Not Find Code (Accepted by IEEE RAL, 2021)

• A Safety Aware Model-Based Reinforcement Learning Framework for Systems with Uncertainties, Paper, Not Find Code (Accepted by American Control Conference, 2021)

• Constrained Reinforcement Learning for Dynamic Optimization under Uncertainty, Paper, Not Find Code (Accepted by IFAC, 2020)

• Benchmarking Safe Deep Reinforcement Learning in Aquatic Navigation, Paper, Not Find Code (Accepted by IROS, 2021)

• Safe Reinforcement Learning with Model Uncertainty Estimates, Paper, Not Find Code (Accepted by IEEE International Conference on Robotics and Automation, 2019)

• Uncertainty-aware Contact-safe Model-based Reinforcement Learning, Paper, Not Find Code (Accepted by IEEE RAL, 2021)

• Uncertainty-Aware Reinforcement Learning for Collision Avoidance, Paper, Not Find Code (Arxiv, 2017)

• Lyapunov-based uncertainty-aware safe reinforcement learning, Paper, Not Find Code (Arxiv, 2021)

• Reinforcement Learning with Uncertainty Estimation for Tactical Decision-Making in Intersections, Paper, Not Find Code (Accepted by IEEE International Conference on Intelligent Transportation Systems, 2020)

• Soft Actor-Critic: Off-Policy Maximum Entropy Deep Reinforcement Learning with a Stochastic Actor, Paper, Code (Accepted by International Conference on Machine Learning, 2018)

• Reinforcement Learning Under Moral Uncertainty, Paper, Code (Accepted by International Conference on Machine Learning, 2021)

• Distributional Reinforcement Learning with Quantile Regression, Paper, Not Find Code (Arxiv, 2019)

• Estimating Risk and Uncertainty in Deep Reinforcement Learning, Paper, Code (Accepted by ICML, 2020)

• Reinforcement Learning for Safety-Critical Control under Model Uncertainty, using Control Lyapunov Functions and Control Barrier Functions, Paper, Not Find Code (Arxiv, 2020)

• Automated Lane Change Decision Making using Deep Reinforcement Learning in Dynamic and Uncertain Highway Environment, Paper, Not Find Code (Accepted by IEEE Intelligent Transportation Systems Conference, 2019)

• Online Robust Reinforcement Learning with Model Uncertainty, Paper, Not Find Code (Accepted by Advances in Neural Information Processing Systems)

• Uncertainty-driven Imagination for Continuous Deep Reinforcement Learning, Paper, Not Find Code (Accepted by Conference on Robot Learning)

• A model for system uncertainty in reinforcement learning, Paper, Not Find Code (Arxiv, 2018)

• Uncertainty-Based Offline Reinforcement Learning with Diversified Q-Ensemble, Paper, Code (Accepted by Advances in Neural Information Processing Systems)

• Reinforcement-Learning-Based Energy Storage System Operation Strategies to Manage Wind Power Forecast Uncertainty, Paper, Not Find Code (Accepted by IEEE Access)

• SENTINEL: Taming Uncertainty with Ensemble based Distributional Reinforcement Learning, Paper, Not Find Code (Accepted by Uncertainty in Artificial Intelligence)

• Exploring State Transition Uncertainty in Variational Reinforcement Learning, Paper, Not Find Code (Accepted by 2020 28th European Signal Processing Conference (EUSIPCO))

Generalization in RL

• Zero-Shot Task Generalization with Multi-Task Deep Reinforcement Learning, Paper, Not Find Code (Accepted by International Conference on Machine Learning, 2017)

• Quantifying Generalization in Reinforcement Learning, Paper, Code (Accepted by International Conference on Machine Learning, 2019)

• NETWORK RANDOMIZATION: A SIMPLE TECHNIQUE FOR GENERALIZATION IN DEEP REINFORCEMENT LEARNING, Paper, Code (Accepted by ICLR, 2020)

• Invariant Policy Optimization: Towards Stronger Generalization in Reinforcement Learning, Paper, Code (Accepted by Learning for Dynamics and Control, 2021)

• Improving Generalization in Reinforcement Learning with Mixture Regularization, Paper, Code (Accepted by Neural Information Processing Systems, 2020)

• High Confidence Generalization for Reinforcement Learning, Paper, Not Find Code (Accepted by International Conference on Machine Learning, 2021)

• Generalization of Reinforcement Learning with Policy-Aware Adversarial Data Augmentation, Paper, Not Find Code (Arxiv, 2021)

• Generalization in Reinforcement Learning: Safely Approximating the Value Function, Paper, Not Find Code (Accepted by Advances in Neural Information Processing Systems(NIPS), 1994)

• Generalization in Reinforcement Learning with Selective Noise Injection and Information Bottleneck, Paper, Code (Accepted by NeurIPS, 2019)

• Generalization in Reinforcement Learning by Soft Data Augmentation, Paper, Code (Accepted by IEEE International Conference on Robotics and Automation (ICRA), 2021)

• Generalization and Regularization in DQN, Paper, Not Find Code (Arxiv, 2020)

• Dynamics Generalization via Information Bottleneck in Deep Reinforcement Learning, Paper, Not Find Code (Arxiv, 2020)

• Domain Randomization for Transferring Deep Neural Networks from Simulation to the Real World, Paper, Not Find Code (Accepted by IEEE International Conference on Intelligent Robots and Systems (IROS), 2017)

• Active Domain Randomization, Paper, Not Find Code (Accepted by Conference on Robot Learning, 2020)