paper-with-me

Papers

OmniSafe: An Infrastructure for Accelerating Safe Reinforcement Learning Research

2023-05-16 · Jiaming Ji, Jiayi Zhou, Borong Zhang, Juntao Dai, Xuehai Pan, Ruiyang Sun, Weidong Huang, Yiran Geng, Mickel Liu, Yaodong Yang

AI systems empowered by reinforcement learning (RL) algorithms harbor the immense potential to catalyze societal advancement, yet their deployment is often impeded by significant safety concerns. Particularly in safety-critical applications, researchers have raised concerns about unintended harms or unsafe behaviors of unaligned RL agents. The philosophy of safe reinforcement learning (SafeRL) is to align RL agents with harmless intentions and safe behavioral patterns. In SafeRL, agents learn to develop optimal policies by receiving feedback from the environment, while also fulfilling the requirement of minimizing the risk of unintended harm or unsafe behavior. However, due to the intricate nature of SafeRL algorithm implementation, combining methodologies across various domains presents a formidable challenge. This had led to an absence of a cohesive and efficacious learning framework within the contemporary SafeRL research milieu. In this work, we introduce a foundational framework designed to expedite SafeRL research endeavors. Our comprehensive framework encompasses an array of algorithms spanning different RL domains and places heavy emphasis on safety elements. Our efforts are to make the SafeRL-related research process more streamlined and efficient, therefore facilitating further research in AI safety. Our project is released at: https://github.com/PKU-Alignment/omnisafe.

📄 PDF Abstract BibTeX arXiv:2305.09304

Code (1)

pku-alignment/omnisafe pytorch

Tasks

Philosophyreinforcement-learningReinforcement LearningReinforcement Learning (RL)Safe Reinforcement Learning

Methods 이 논문이 사용한 방법론

ALIGN In the ALIGN method, visual and language representations are jointly trained from noisy image alt-text data. The image and text encoders are learned via contrastive loss…

Similar Papers 제목 키워드 기반

OmniSafeBench-MM: A Unified Benchmark and Toolbox for Multimodal Jailbreak Attack-Defense Evaluation

2025-12-06 · Xiaojun Jia, Jie Liao, Qi Guo, Teng Ma 외 arxiv

Recent advances in multi-modal large language models (MLLMs) have enabled unified perception-reasoning capabilities, yet these systems remain highly vulnerable to jailbreak attacks that bypass safety alignment and induce…

SafeOR-Gym: A Benchmark Suite for Safe Reinforcement Learning Algorithms on Practical Operations Research Problems

2025-06-02 · Asha Ramanujam, Adam Elyoumi, Hao Chen, Sai Madhukiran Kompalli 외 arxiv

Most existing safe reinforcement learning (RL) benchmarks focus on robotics and control tasks, offering limited relevance to high-stakes domains that involve structured constraints, mixed-integer decisions, and industria…

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 외

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 challe…

MuJoCoReinforcement Learning (RL)Safe Reinforcement Learning

Enhancing Efficiency of Safe Reinforcement Learning via Sample Manipulation

2024-05-31 · Shangding Gu, Laixi Shi, Yuhao Ding, Alois Knoll 외

Safe reinforcement learning (RL) is crucial for deploying RL agents in real-world applications, as it aims to maximize long-term rewards while satisfying safety constraints. However, safe RL often suffers from sample ine…

MuJoCoreinforcement-learningReinforcement LearningReinforcement Learning (RL)+1

KFCPO: Kronecker-Factored Approximated Constrained Policy Optimization

2025-11-02 · Joonyoung Lim, Younghwan Yoo arxiv

We propose KFCPO, a novel Safe Reinforcement Learning (Safe RL) algorithm that combines scalable Kronecker-Factored Approximate Curvature (K-FAC) based second-order policy optimization with safety-aware gradient manipula…

Reinforcement Learning