Papers › Robust Lagrangian and Adversarial Policy Gradient for Robust Constrained Markov...

Robust Lagrangian and Adversarial Policy Gradient for Robust Constrained Markov Decision Processes

22 Aug 2023arXiv:2308.11267archive 2025-07-28

David M. Bossens

The robust constrained Markov decision process (RCMDP) is a recent task-modelling framework for reinforcement learning that incorporates behavioural constraints and that provides robustness to errors in the transition dynamics model through the use of an uncertainty set. Simulating RCMDPs requires computing the worst-case dynamics based on value estimates for each state, an approach which has previously been used in the Robust Constrained Policy Gradient (RCPG). Highlighting potential downsides of RCPG such as not robustifying the full constrained objective and the lack of incremental learning, this paper introduces two algorithms, called RCPG with Robust Lagrangian and Adversarial RCPG. RCPG with Robust Lagrangian modifies RCPG by taking the worst-case dynamics based on the Lagrangian rather than either the value or the constraint. Adversarial RCPG also formulates the worst-case dynamics based on the Lagrangian but learns this directly and incrementally as an adversarial policy through gradient descent rather than indirectly and abruptly through constrained optimisation on a sorted value list. A theoretical analysis first derives the Lagrangian policy gradient for the policy optimisation of both proposed algorithms and then the adversarial policy gradient to learn the adversary for Adversarial RCPG. Empirical experiments injecting perturbations in inventory management and safe navigation tasks demonstrate the competitive performance of both algorithms compared to traditional RCPG variants as well as non-robust and non-constrained ablations. In particular, Adversarial RCPG ranks among the top two performing algorithms on all tests.

PaperPDFCode

Code

bossdm/RCMDP officialmentioned on GitHubpytorch report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Incremental Learning

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections