Copertina

Escaping Offline Pessimism: Vector-Field Reward Shaping for Safe Frontier Exploration

Fuente: arXiv

Salvato in:

Dettagli Bibliografici
Autori principali:	Roknilamouki, Amirhossein, Ghosh, Arnob, Ekici, Eylem, Shroff, Ness B.
Natura:	Preprint
Pubblicazione:	2026
Soggetti:	Machine Learning
Accesso online:	Acceder al recurso
Tags:	Aggiungi Tag Nessun Tag, puoi essere il primo ad aggiungerne!!

Documenti analoghi

Provably Efficient RL for Linear MDPs under Instantaneous Safety Constraints in Non-Convex Feature Spaces
di: Roknilamouki, Amirhossein, et al.
Pubblicazione: (2025)

FIRM: Federated In-client Regularized Multi-objective Alignment for Large Language Models
di: Nourzad, Fatemeh, et al.
Pubblicazione: (2025)

Online Learning for Optimizing AoI-Energy Tradeoff under Unknown Channel Statistics
di: Abd-Elmagid, Mohamed A., et al.
Pubblicazione: (2025)

Beyond Freshness and Semantics: A Coupon-Collector Framework for Effective Status Updates
di: Ahmed, Youssef, et al.
Pubblicazione: (2026)

Adversarially Trained Weighted Actor-Critic for Safe Offline Reinforcement Learning
di: Wei, Honghao, et al.
Pubblicazione: (2024)

Provable Last-Iterate Convergence for Multi-Objective Safe LLM Alignment via Optimistic Primal-Dual
di: Li, Yining, et al.
Pubblicazione: (2026)

How to Find the Exact Pareto Front for Multi-Objective MDPs?
di: Li, Yining, et al.
Pubblicazione: (2024)

Provably Efficient Multi-Objective Bandit Algorithms under Preference-Centric Customization
di: Cao, Linfeng, et al.
Pubblicazione: (2025)

Improved Regret Bound for Safe Reinforcement Learning via Tighter Cost Pessimism and Reward Optimism
di: Yu, Kihyun, et al.
Pubblicazione: (2024)

From Scores to Gibbs Correctors: Accelerating Uniform-Rate Discrete Diffusion Models
di: Liang, Yuchen, et al.
Pubblicazione: (2026)

From Robotics to Sepsis Treatment: Offline RL via Geometric Pessimism
di: Wanjari, Sarthak
Pubblicazione: (2026)

A Tale of Two Cities: Pessimism and Opportunism in Offline Dynamic Pricing
di: Bian, Zeyu, et al.
Pubblicazione: (2024)

Beyond Pessimism: Offline Learning in KL-regularized Games
di: Zhang, Yuheng, et al.
Pubblicazione: (2026)

Monitoring State Transitions in Markovian Systems with Sampling Cost
di: Saurav, Kumar, et al.
Pubblicazione: (2025)

Near-Optimal Partially Observable Reinforcement Learning with Partial Online State Information
di: Shi, Ming, et al.
Pubblicazione: (2023)

Tail-Risk-Safe Monte Carlo Tree Search under PAC-Level Guarantees
di: Zhang, Zuyuan, et al.
Pubblicazione: (2025)

From Curiosity to Caution: Mitigating Reward Hacking for Best-of-N with Pessimism
di: Yu, Zhuohao, et al.
Pubblicazione: (2026)

Pessimism Meets Risk: Risk-Sensitive Offline Reinforcement Learning
di: Zhang, Dake, et al.
Pubblicazione: (2024)

Regret Bounds for Reinforcement Learning from Multi-Source Imperfect Preferences
di: Shi, Ming, et al.
Pubblicazione: (2026)

BeST -- A Novel Source Selection Metric for Transfer Learning
di: Soni, Ashutosh, et al.
Pubblicazione: (2025)

Sharp Convergence Rates for Masked Diffusion Models
di: Liang, Yuchen, et al.
Pubblicazione: (2026)

Theory on Score-Mismatched Diffusion Models and Zero-Shot Conditional Samplers
di: Liang, Yuchen, et al.
Pubblicazione: (2024)

An LP-based Sampling Policy for Multi-Armed Bandits with Side-Observations and Stochastic Availability
di: Soni, Ashutosh, et al.
Pubblicazione: (2026)

Unified PAC-Bayesian Study of Pessimism for Offline Policy Learning with Regularized Importance Sampling
di: Aouali, Imad, et al.
Pubblicazione: (2024)

Pessimism-Free Offline Learning in General-Sum Games via KL Regularization
di: Chen, Claire, et al.
Pubblicazione: (2026)

LexiSafe: Offline Safe Reinforcement Learning with Lexicographic Safety-Reward Hierarchy
di: Yang, Hsin-Jung, et al.
Pubblicazione: (2026)

Discrete Diffusion Models: Novel Analysis and New Sampler Guarantees
di: Liang, Yuchen, et al.
Pubblicazione: (2025)

Broadening Target Distributions for Accelerated Diffusion Models via a Novel Analysis Approach
di: Liang, Yuchen, et al.
Pubblicazione: (2024)

Mixture-of-Transformers Learn Faster: A Theoretical Study on Classification Problems
di: Li, Hongbo, et al.
Pubblicazione: (2025)

OASIS: Conditional Distribution Shaping for Offline Safe Reinforcement Learning
di: Yao, Yihang, et al.
Pubblicazione: (2024)

Provably Efficient Sample Complexity for Robust CMDP
di: Ganguly, Sourav, et al.
Pubblicazione: (2025)

Prediction-Assisted Online Distributed Deep Learning Workload Scheduling in GPU Clusters
di: Luo, Ziyue, et al.
Pubblicazione: (2025)

Automatic Reward Shaping from Confounded Offline Data
di: Li, Mingxuan, et al.
Pubblicazione: (2025)

Can We Theoretically Quantify the Impacts of Local Updates on the Generalization Performance of Federated Learning?
di: Ju, Peizhong, et al.
Pubblicazione: (2024)

The Virtues of Pessimism in Inverse Reinforcement Learning
di: Wu, David, et al.
Pubblicazione: (2024)

Constraint-Rectified Training for Efficient Chain-of-Thought
di: Wu, Qinhang, et al.
Pubblicazione: (2026)

Theory on Mixture-of-Experts in Continual Learning
di: Li, Hongbo, et al.
Pubblicazione: (2024)

Certifiable Safe RLHF: Fixed-Penalty Constraint Optimization for Safer Language Models
di: Pandit, Kartik, et al.
Pubblicazione: (2025)

Sparse-Reg: Improving Sample Complexity in Offline Reinforcement Learning using Sparsity
di: Arnob, Samin Yeasar, et al.
Pubblicazione: (2025)

GPT-4o as the Gold Standard: A Scalable and General Purpose Approach to Filter Language Model Pretraining Data
di: Zhang, Jifan, et al.
Pubblicazione: (2024)