paper-with-me

Papers

Balance Reward and Safety Optimization for Safe Reinforcement Learning: A Perspective of Gradient Manipulation

2024-05-02 · Shangding Gu, Bilgehan Sel, Yuhao Ding, Lu Wang, QIngwei Lin, Ming Jin, Alois Knoll

Ensuring the safety of Reinforcement Learning (RL) is crucial for its deployment in real-world applications. Nevertheless, managing the trade-off between reward and safety during exploration presents a significant challenge. Improving reward performance through policy adjustments may adversely affect safety performance. In this study, we aim to address this conflicting relation by leveraging the theory of gradient manipulation. Initially, we analyze the conflict between reward and safety gradients. Subsequently, we tackle the balance between reward and safety optimization by proposing a soft switching policy optimization method, for which we provide convergence analysis. Based on our theoretical examination, we provide a safe RL framework to overcome the aforementioned challenge, and we develop a Safety-MuJoCo Benchmark to assess the performance of safe RL algorithms. Finally, we evaluate the effectiveness of our method on the Safety-MuJoCo Benchmark and a popular safe RL benchmark, Omnisafe. Experimental results demonstrate that our algorithms outperform several state-of-the-art baselines in terms of balancing reward and safety optimization.

📄 PDF Abstract BibTeX arXiv:2405.01677

Code (3)

pku-alignment/omnisafe 공식 구현 pytorch
saferl-lab/safety-mujoco 공식 구현
SafeRL-Lab/Safe-Multi-Objective-MuJoCo

Tasks

MuJoCoReinforcement Learning (RL)Safe Reinforcement Learning

Similar Papers 제목 키워드 기반

SCPO: Safe Reinforcement Learning with Safety Critic Policy Optimization

2023-11-01 · Jaafar Mhamed, Shangding Gu

Incorporating safety is an essential prerequisite for broadening the practical applications of reinforcement learning in real-world scenarios. To tackle this challenge, Constrained Markov Decision Processes (CMDPs) are l…

Benchmarkingreinforcement-learningReinforcement LearningSafe Reinforcement Learning

SB-TRPO: Towards Safe Reinforcement Learning with Hard Constraints

2025-12-29 · Dominik Wagner, Ankit Kanwar, Luke Ong arxiv

In safety-critical domains, reinforcement learning (RL) agents must often satisfy strict, zero-cost safety constraints while accomplishing tasks. Existing model-free methods frequently either fail to achieve near-zero sa…

Reinforcement Learning

LexiSafe: Offline Safe Reinforcement Learning with Lexicographic Safety-Reward Hierarchy

2026-02-19 · Hsin-Jung Yang, Zhanhong Jiang, Prajwal Koirala, Qisai Liu 외 arxiv

Offline safe reinforcement learning (RL) is increasingly important for cyber-physical systems (CPS), where safety violations during training are unacceptable and only pre-collected data are available. Existing offline sa…

Reinforcement LearningOffline RL

Safe CoR: A Dual-Expert Approach to Integrating Imitation Learning and Safe Reinforcement Learning Using Constraint Rewards

2024-07-02 · Hyeokjin Kwon, Gunmin Lee, Junseo Lee, Songhwai Oh

In the realm of autonomous agents, ensuring safety and reliability in complex and dynamic environments remains a paramount challenge. Safe reinforcement learning addresses these concerns by introducing safety constraints…

Imitation LearningSafe Reinforcement Learning

HAIR: Hardness-Aware Inverse Reinforcement Learning with Introspective Reasoning for LLM Alignment

2025-03-23 · Ruoxi Cheng, Haoxuan Ma, Weixin Wang

The alignment of large language models (LLMs) with human values remains critical yet hindered by four key challenges: (1) scarcity of balanced safety datasets, (2) alignment tax, (3) vulnerability to jailbreak attacks du…