The Curious Price of Distributional Robustness in Reinforcement Learning with a Generative Model

• URL: http://arxiv.org/abs/2305.16589v2
• Date: Fri, 12 Apr 2024 08:09:33 GMT
• Title: The Curious Price of Distributional Robustness in Reinforcement Learning with a Generative Model
• Authors: Laixi Shi, Gen Li, Yuting Wei, Yuxin Chen, Matthieu Geist, Yuejie Chi,
• Abstract summary: This paper investigates model robustness in reinforcement learning (RL) to reduce the sim-to-real gap in practice. We adopt the framework of distributionally robust Markov decision processes (RMDPs), aimed at learning a policy that optimize the worst-case performance when the deployed environment falls within a prescribed uncertainty set around the nominal MDP.
• Score: 61.87673435273466
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: This paper investigates model robustness in reinforcement learning (RL) to reduce the sim-to-real gap in practice. We adopt the framework of distributionally robust Markov decision processes (RMDPs), aimed at learning a policy that optimizes the worst-case performance when the deployed environment falls within a prescribed uncertainty set around the nominal MDP. Despite recent efforts, the sample complexity of RMDPs remained mostly unsettled regardless of the uncertainty set in use. It was unclear if distributional robustness bears any statistical consequences when benchmarked against standard RL. Assuming access to a generative model that draws samples based on the nominal MDP, we characterize the sample complexity of RMDPs when the uncertainty set is specified via either the total variation (TV) distance or $\chi^2$ divergence. The algorithm studied here is a model-based method called {\em distributionally robust value iteration}, which is shown to be near-optimal for the full range of uncertainty levels. Somewhat surprisingly, our results uncover that RMDPs are not necessarily easier or harder to learn than standard MDPs. The statistical consequence incurred by the robustness requirement depends heavily on the size and shape of the uncertainty set: in the case w.r.t.~the TV distance, the minimax sample complexity of RMDPs is always smaller than that of standard MDPs; in the case w.r.t.~the $\chi^2$ divergence, the sample complexity of RMDPs can often far exceed the standard MDP counterpart.

Related papers

• Distributionally Robust Reinforcement Learning with Interactive Data Collection: Fundamental Hardness and Near-Optimal Algorithm [14.517103323409307]
  Sim-to-real gap represents disparity between training and testing environments. A promising approach to addressing this challenge is distributionally robust RL. We tackle robust RL via interactive data collection and present an algorithm with a provable sample complexity guarantee.
  arXiv  Detail & Related papers  (2024-04-04T16:40:22Z)
• Sample Complexity of Offline Distributionally Robust Linear Markov Decision Processes [37.15580574143281]
  offline reinforcement learning (RL) This paper considers the sample complexity of distributionally robust linear Markov decision processes (MDPs) with an uncertainty set characterized by the total variation distance using offline data. We develop a pessimistic model-based algorithm and establish its sample complexity bound under minimal data coverage assumptions.
  arXiv  Detail & Related papers  (2024-03-19T17:48:42Z)
• Provably Efficient Algorithm for Nonstationary Low-Rank MDPs [48.92657638730582]
  We make the first effort to investigate nonstationary RL under episodic low-rank MDPs, where both transition kernels and rewards may vary over time. We propose a parameter-dependent policy optimization algorithm called PORTAL, and further improve PORTAL to its parameter-free version of Ada-PORTAL. For both algorithms, we provide upper bounds on the average dynamic suboptimality gap, which show that as long as the nonstationarity is not significantly large, PORTAL and Ada-PORTAL are sample-efficient and can achieve arbitrarily small average dynamic suboptimality gap with sample complexity.
  arXiv  Detail & Related papers  (2023-08-10T09:52:44Z)
• Non-stationary Reinforcement Learning under General Function Approximation [60.430936031067006]
  We first propose a new complexity metric called dynamic Bellman Eluder (DBE) dimension for non-stationary MDPs. Based on the proposed complexity metric, we propose a novel confidence-set based model-free algorithm called SW-OPEA. We show that SW-OPEA is provably efficient as long as the variation budget is not significantly large.
  arXiv  Detail & Related papers  (2023-06-01T16:19:37Z)
• Twice Regularized Markov Decision Processes: The Equivalence between Robustness and Regularization [64.60253456266872]
  Markov decision processes (MDPs) aim to handle changing or partially known system dynamics. Regularized MDPs show more stability in policy learning without impairing time complexity. Bellman operators enable us to derive planning and learning schemes with convergence and generalization guarantees.
  arXiv  Detail & Related papers  (2023-03-12T13:03:28Z)
• Distributionally Robust Model-Based Offline Reinforcement Learning with Near-Optimal Sample Complexity [39.886149789339335]
  offline reinforcement learning aims to learn to perform decision making from history data without active exploration. Due to uncertainties and variabilities of the environment, it is critical to learn a robust policy that performs well even when the deployed environment deviates from the nominal one used to collect the history dataset. We consider a distributionally robust formulation of offline RL, focusing on robust Markov decision processes with an uncertainty set specified by the Kullback-Leibler divergence in both finite-horizon and infinite-horizon settings.
  arXiv  Detail & Related papers  (2022-08-11T11:55:31Z)
• Robust Entropy-regularized Markov Decision Processes [23.719568076996662]
  We study a robust version of the ER-MDP model, where the optimal policies are required to be robust. We show that essential properties that hold for the non-robust ER-MDP and robust unregularized MDP models also hold in our settings. We show how our framework and results can be integrated into different algorithmic schemes including value or (modified) policy.
  arXiv  Detail & Related papers  (2021-12-31T09:50:46Z)
• Sample Complexity of Robust Reinforcement Learning with a Generative Model [0.0]
  We propose a model-based reinforcement learning (RL) algorithm for learning an $epsilon$-optimal robust policy. We consider three different forms of uncertainty sets, characterized by the total variation distance, chi-square divergence, and KL divergence. In addition to the sample complexity results, we also present a formal analytical argument on the benefit of using robust policies.
  arXiv  Detail & Related papers  (2021-12-02T18:55:51Z)
• Twice regularized MDPs and the equivalence between robustness and regularization [65.58188361659073]
  We show that policy iteration on reward-robust MDPs can have the same time complexity as on regularized MDPs. We generalize regularized MDPs to twice regularized MDPs.
  arXiv  Detail & Related papers  (2021-10-12T18:33:45Z)
• Breaking the Sample Size Barrier in Model-Based Reinforcement Learning with a Generative Model [50.38446482252857]
  This paper is concerned with the sample efficiency of reinforcement learning, assuming access to a generative model (or simulator) We first consider $gamma$-discounted infinite-horizon Markov decision processes (MDPs) with state space $mathcalS$ and action space $mathcalA$. We prove that a plain model-based planning algorithm suffices to achieve minimax-optimal sample complexity given any target accuracy level.
  arXiv  Detail & Related papers  (2020-05-26T17:53:18Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.