Image de couverture de livre

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

Fuente: arXiv

Enregistré dans:

Détails bibliographiques
Auteurs principaux:	Gu, Shangding, Sel, Bilgehan, Ding, Yuhao, Wang, Lu, Lin, Qingwei, Jin, Ming, Knoll, Alois
Format:	Preprint
Publié:	2024
Sujets:	Machine Learning Artificial Intelligence
Accès en ligne:	Acceder al recurso
Tags:	Ajouter un tag Pas de tags, Soyez le premier à ajouter un tag!

Documents similaires

Safe and Balanced: A Framework for Constrained Multi-Objective Reinforcement Learning
par: Gu, Shangding, et autres
Publié: (2024)

Enhancing Efficiency of Safe Reinforcement Learning via Sample Manipulation
par: Gu, Shangding, et autres
Publié: (2024)

A Review of Safe Reinforcement Learning: Methods, Theory and Applications
par: Gu, Shangding, et autres
Publié: (2022)

LLMs Can Plan Only If We Tell Them
par: Sel, Bilgehan, et autres
Publié: (2025)

Safe Continual Domain Adaptation after Sim2Real Transfer of Reinforcement Learning Policies in Robotics
par: Josifovski, Josip, et autres
Publié: (2025)

TeaMs-RL: Teaching LLMs to Generate Better Instruction Datasets via Reinforcement Learning
par: Gu, Shangding, et autres
Publié: (2024)

A CMDP-within-online framework for Meta-Safe Reinforcement Learning
par: Khattar, Vanshaj, et autres
Publié: (2024)

Reinforcement Learning with Backtracking Feedback
par: Sel, Bilgehan, et autres
Publié: (2026)

Backtracking for Safety
par: Sel, Bilgehan, et autres
Publié: (2025)

Trojan-Speak: Bypassing Constitutional Classifiers with No Jailbreak Tax via Adversarial Finetuning
par: Sel, Bilgehan, et autres
Publié: (2026)

Algorithm of Thoughts: Enhancing Exploration of Ideas in Large Language Models
par: Sel, Bilgehan, et autres
Publié: (2023)

Multi-Agent Reinforcement Learning for Autonomous Driving: A Survey
par: Zhang, Ruiqi, et autres
Publié: (2024)

Long Context, Less Focus: A Scaling Gap in LLMs Revealed through Privacy and Personalization
par: Gu, Shangding
Publié: (2026)

Remembering More, Risking More: Longitudinal Safety Risks in Memory-Equipped LLM Agents
par: Al-Tawaha, Ahmad, et autres
Publié: (2026)

Skin-in-the-Game: Decision Making via Multi-Stakeholder Alignment in LLMs
par: Sel, Bilgehan, et autres
Publié: (2024)

Generating Automotive Code: Large Language Models for Software Development and Verification in Safety-Critical Systems
par: Kirchner, Sven, et autres
Publié: (2025)

A New Perspective On AI Safety Through Control Theory Methodologies
par: Ullrich, Lars, et autres
Publié: (2025)

Analysis of Randomization Effects on Sim2Real Transfer in Reinforcement Learning for Robotic Manipulation Tasks
par: Josifovski, Josip, et autres
Publié: (2022)

Robust Gymnasium: A Unified Modular Benchmark for Robust Reinforcement Learning
par: Gu, Shangding, et autres
Publié: (2025)

Safe Multi-Agent Reinforcement Learning with Bilevel Optimization in Autonomous Driving
par: Zheng, Zhi, et autres
Publié: (2024)

Data Uniformity Improves Training Efficiency and More, with a Convergence Framework Beyond the NTK Regime
par: Wang, Yuqing, et autres
Publié: (2025)

LLM-Empowered Functional Safety and Security by Design in Automotive Systems
par: Petrovic, Nenad, et autres
Publié: (2026)

Safety-Aligned 3D Object Detection: Single-Vehicle, Cooperative, and End-to-End Perspectives
par: Liao, Brian Hsuan-Cheng, et autres
Publié: (2026)

State Representations as Incentives for Reinforcement Learning Agents: A Sim2Real Analysis on Robotic Grasping
par: Petropoulakis, Panagiotis, et autres
Publié: (2023)

StyleBench: Evaluating thinking styles in Large Language Models
par: Guo, Junyu, et autres
Publié: (2025)

LLMs Should Express Uncertainty Explicitly
par: Guo, Junyu, et autres
Publié: (2026)

Enhancing Safe and Controllable Protein Generation via Knowledge Preference Optimization
par: Wang, Yuhao, et autres
Publié: (2025)

AgenticPay: A Multi-Agent LLM Negotiation System for Buyer-Seller Transactions
par: Liu, Xianyang, et autres
Publié: (2026)

Pretrain Value, Not Reward: Decoupled Value Policy Optimization
par: Huang, Chenghua, et autres
Publié: (2025)

Self-Evolved Reward Learning for LLMs
par: Huang, Chenghua, et autres
Publié: (2024)

Trapezoidal Gradient Descent for Effective Reinforcement Learning in Spiking Networks
par: Pan, Yuhao, et autres
Publié: (2024)

Position: AI Safety Must Embrace an Antifragile Perspective
par: Jin, Ming, et autres
Publié: (2025)

Balancing Rewards in Text Summarization: Multi-Objective Reinforcement Learning via HyperVolume Optimization
par: Song, Junjie, et autres
Publié: (2025)

Constraint-Conditioned Policy Optimization for Versatile Safe Reinforcement Learning
par: Yao, Yihang, et autres
Publié: (2023)

Safe Reinforcement Learning with Learned Non-Markovian Safety Constraints
par: Low, Siow Meng, et autres
Publié: (2024)

Reinforcement Learning in a Safety-Embedded MDP with Trajectory Optimization
par: Yang, Fan, et autres
Publié: (2023)

Leveraging Analytic Gradients in Provably Safe Reinforcement Learning
par: Walter, Tim, et autres
Publié: (2025)

Pretrained Bayesian Non-parametric Knowledge Prior in Robotic Long-Horizon Reinforcement Learning
par: Meng, Yuan, et autres
Publié: (2025)

Safety Modulation: Enhancing Safety in Reinforcement Learning through Cost-Modulated Rewards
par: Zhang, Hanping, et autres
Publié: (2025)

Safe CoR: A Dual-Expert Approach to Integrating Imitation Learning and Safe Reinforcement Learning Using Constraint Rewards
par: Kwon, Hyeokjin, et autres
Publié: (2024)