Twice Regularized Markov Decision Processes: The Equivalence between Robustness and Regularization
Preprint
Abstract
Robust Markov decision processes (MDPs) aim to handle changing or partially known system dynamics. To solve them, one typically resorts to robust optimization methods. However, this significantly increases computational complexity and limits scalability in both learning and planning. On the other hand, regularized MDPs show more stability in policy learning without impairing time complexity. Yet, they generally do not encompass uncertainty in the model dynamics. In this work, we aim to learn robust MDPs using regularization. We first show that regularized MDPs are a particular instance of robust MDPs with uncertain reward. We thus establish that policy iteration on reward-robust MDPs can have the same time complexity as on regularized MDPs. We further extend this relationship to MDPs with uncertain transitions: this leads to a regularization term with an additional dependence on the value function. We then generalize regularized MDPs to twice regularized MDPs ($\text{R}^2$ MDPs), i.e., MDPs with $\textit{both}$ value and policy regularization. The corresponding Bellman operators enable us to derive planning and learning schemes with convergence and generalization guarantees, thus reducing robustness to regularization. We numerically show this two-fold advantage on tabular and physical domains, highlighting the fact that $\text{R}^2$ preserves its efficacy in continuous environments.
Keywords
All Related Versions
This publication has 8 references indexed in Scilit:
- Robust Markov Decision Processes: Beyond RectangularityMathematics of Operations Research, 2023
- Fast Global Convergence of Natural Policy Gradient Methods with Entropy RegularizationOperations Research, 2022
- Robust Value Iteration for Continuous Control Tasks2021
- Sparse Markov Decision Processes With Causal Sparse Tsallis Entropy Regularization for Reinforcement LearningIEEE Robotics and Automation Letters, 2018
- Robust MDPs with k-Rectangular UncertaintyMathematics of Operations Research, 2016
- Bias and Variance Approximation in Value Function EstimatesManagement Science, 2007
- Robust Control of Markov Decision Processes with Uncertain Transition MatricesOperations Research, 2005
- On the Convergence of Stochastic Iterative Dynamic Programming AlgorithmsNeural Computation, 1994