Primal-Only Actor Critic Algorithm for Robust Constrained Average Cost MDPs

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Satheesh, Anirudh, Sathish, Sooraj, Ganesh, Swetha, Powell, Keenan, Aggarwal, Vaneet
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917069245644800
author Satheesh, Anirudh
Sathish, Sooraj
Ganesh, Swetha
Powell, Keenan
Aggarwal, Vaneet
author_facet Satheesh, Anirudh
Sathish, Sooraj
Ganesh, Swetha
Powell, Keenan
Aggarwal, Vaneet
contents In this work, we study the problem of finding robust and safe policies in Robust Constrained Average-Cost Markov Decision Processes (RCMDPs). A key challenge in this setting is the lack of strong duality, which prevents the direct use of standard primal-dual methods for constrained RL. Additional difficulties arise from the average-cost setting, where the Robust Bellman operator is not a contraction under any norm. To address these challenges, we propose an actor-critic algorithm for Average-Cost RCMDPs. We show that our method achieves both \(ε\)-feasibility and \(ε\)-optimality, and we establish a sample complexities of \(\tilde{O}\left(ε^{-4}\right)\) and \(\tilde{O}\left(ε^{-6}\right)\) with and without slackness assumption, which is comparable to the discounted setting.
format Preprint
id arxiv_https___arxiv_org_abs_2511_05758
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Primal-Only Actor Critic Algorithm for Robust Constrained Average Cost MDPs
Satheesh, Anirudh
Sathish, Sooraj
Ganesh, Swetha
Powell, Keenan
Aggarwal, Vaneet
Machine Learning
In this work, we study the problem of finding robust and safe policies in Robust Constrained Average-Cost Markov Decision Processes (RCMDPs). A key challenge in this setting is the lack of strong duality, which prevents the direct use of standard primal-dual methods for constrained RL. Additional difficulties arise from the average-cost setting, where the Robust Bellman operator is not a contraction under any norm. To address these challenges, we propose an actor-critic algorithm for Average-Cost RCMDPs. We show that our method achieves both \(ε\)-feasibility and \(ε\)-optimality, and we establish a sample complexities of \(\tilde{O}\left(ε^{-4}\right)\) and \(\tilde{O}\left(ε^{-6}\right)\) with and without slackness assumption, which is comparable to the discounted setting.
title Primal-Only Actor Critic Algorithm for Robust Constrained Average Cost MDPs
topic Machine Learning
url https://arxiv.org/abs/2511.05758