Lyapunov Robust Constrained-mdps: Soft-constrained Robustly Stable Policy Optimization Under Model Uncertainty

Abstract

Safety and robustness are two desired properties for any reinforcement learning algorithm. CMDPs can handle additional safety constraints and RMDPs can perform well under model uncertainties. In this paper, we propose to unite these two frameworks resulting in robust constrained MDPs (RCMDPs). The motivation is to develop a framework that can satisfy safety constraints while also simultaneously offer robustness to model uncertainties. We develop the RCMDP objective, derive gradient update formula to optimize this objective and then propose policy gradient based algorithms. We also independently propose Lyapunov based reward shaping for RCMDPs, yielding better stability and convergence properties.

Stats

• citations0
• S2 citations—
• github stars0
• HF likes0
• heat score0.00
• arxiv keyrussel2021lyapunov

Related papers

• Robust Lagrangian And Adversarial Policy Gradient For Robust Constrained Markov Decision Processes (2023)2.26
• Efficient Policy Optimization In Robust Constrained Mdps With Iteration Complexity Guarantees (2025)0.00
• Robust Reinforcement Learning Using Least Squares Policy Iteration With Provable Performance Guarantees (2020)0.00
• Solving Non-rectangular Reward-robust Mdps Via Frequency Regularization (2023)0.00
• Robust Model-based Reinforcement Learning With An Adversarial Auxiliary Model (2024)0.00
• Sample Complexity Of Robust Reinforcement Learning With A Generative Model (2021)0.00
• A Policy Gradient Primal-dual Algorithm For Constrained Mdps With Uniform PAC Guarantees (2024)0.00
• Tight Bayesian Ambiguity Sets For Robust Mdps (2018)0.00