arXiv · 2604.21177
Revisiting Subgradient Dominance in Robust MDPs: Counterexamples, Hardness, and Sufficient Conditions
Abstract
Projected subgradient descent (PSD) has gained popularity for solving robust Markov decision processes (RMDPs) because it applies to a broader class of uncertainty sets than traditional dynamic programming. Existing work claims that RMDPs with a general compact uncertainty set satisfy the subgradient dominance property, under which exact PSD converges to an $\varepsilon$-optimal policy in a polynomial number of updates (e.g., Wang et al., 2023). We show that these claims are incorrect. Even when the uncertainty set has cardinality two, the RMDP objective is not subgradient-dominant and can admit suboptimal strict local minima. Moreover, we prove that finding an $\varepsilon$-optimal policy can be NP-hard even in settings where subgradients are efficiently computable: (i) finite transition uncertainty sets and (ii) $sa$-rectangular finite transition uncertainty sets with finite cost uncertainty sets. Finally, we identify two conditions under which RMDPs do satisfy subgradient dominance: when, for each policy, either the worst-case transition kernel or the worst-case action-value function is unique.
Explore related subjects
Keep this discovery
Explore connections, maps & timelines
Toshinori Kitamura, Arnob Ghosh, Alex Ayoub, Thang D. Chu, Csaba Szepesvári. 2026-04-23. Revisiting Subgradient Dominance in Robust MDPs: Counterexamples, Hardness, and Sufficient Conditions. https://arxiv.org/abs/2604.21177
Cite the original work for its findings. Save a collection to share your selection of sources.