Enregistré dans:
Détails bibliographiques
Auteurs principaux: Ma, Shaocong, Chen, Ziyi, Zhou, Yi, Huang, Heng
Format: Preprint
Publié: 2025
Sujets:
Accès en ligne:https://arxiv.org/abs/2508.17448
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866918144687210496
author Ma, Shaocong
Chen, Ziyi
Zhou, Yi
Huang, Heng
author_facet Ma, Shaocong
Chen, Ziyi
Zhou, Yi
Huang, Heng
contents The goal of robust constrained reinforcement learning (RL) is to optimize an agent's performance under the worst-case model uncertainty while satisfying safety or resource constraints. In this paper, we demonstrate that strong duality does not generally hold in robust constrained RL, indicating that traditional primal-dual methods may fail to find optimal feasible policies. To overcome this limitation, we propose a novel primal-only algorithm called Rectified Robust Policy Optimization (RRPO), which operates directly on the primal problem without relying on dual formulations. We provide theoretical convergence guarantees under mild regularity assumptions, showing convergence to an approximately optimal feasible policy with iteration complexity matching the best-known lower bound when the uncertainty set diameter is controlled in a specific level. Empirical results in a grid-world environment validate the effectiveness of our approach, demonstrating that RRPO achieves robust and safe performance under model uncertainties while the non-robust method can violate the worst-case safety constraints.
format Preprint
id arxiv_https___arxiv_org_abs_2508_17448
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Rectified Robust Policy Optimization for Model-Uncertain Constrained Reinforcement Learning without Strong Duality
Ma, Shaocong
Chen, Ziyi
Zhou, Yi
Huang, Heng
Machine Learning
The goal of robust constrained reinforcement learning (RL) is to optimize an agent's performance under the worst-case model uncertainty while satisfying safety or resource constraints. In this paper, we demonstrate that strong duality does not generally hold in robust constrained RL, indicating that traditional primal-dual methods may fail to find optimal feasible policies. To overcome this limitation, we propose a novel primal-only algorithm called Rectified Robust Policy Optimization (RRPO), which operates directly on the primal problem without relying on dual formulations. We provide theoretical convergence guarantees under mild regularity assumptions, showing convergence to an approximately optimal feasible policy with iteration complexity matching the best-known lower bound when the uncertainty set diameter is controlled in a specific level. Empirical results in a grid-world environment validate the effectiveness of our approach, demonstrating that RRPO achieves robust and safe performance under model uncertainties while the non-robust method can violate the worst-case safety constraints.
title Rectified Robust Policy Optimization for Model-Uncertain Constrained Reinforcement Learning without Strong Duality
topic Machine Learning
url https://arxiv.org/abs/2508.17448