Wasserstein Distributionally Robust Stochastic Control: A Data-Driven Approach
From MaRDI portal
Abstract: Standard stochastic control methods assume that the probability distribution of uncertain variables is available. Unfortunately, in practice, obtaining accurate distribution information is a challenging task. To resolve this issue, we investigate the problem of designing a control policy that is robust against errors in the empirical distribution obtained from data. This problem can be formulated as a two-player zero-sum dynamic game problem, where the action space of the adversarial player is a Wasserstein ball centered at the empirical distribution. We propose computationally tractable value and policy iteration algorithms with explicit estimates of the number of iterations required for constructing an -optimal policy. We show that the contraction property of associated Bellman operators extends a single-stage out-of-sample performance guarantee, obtained using a measure concentration inequality, to the corresponding multi-stage guarantee without any degradation in the confidence level. In addition, we characterize an explicit form of the optimal distributionally robust control policy and the worst-case distribution policy for linear-quadratic problems with Wasserstein penalty. Our study indicates that dynamic programming and Kantorovich duality play a critical role in solving and analyzing the Wasserstein distributionally robust stochastic control problems.
Recommendations
- Distributional Robustness in Minimax Linear Quadratic Control with Wasserstein Distance
- Data-driven distributionally robust optimization using the Wasserstein metric: performance guarantees and tractable reformulations
- Data-driven distributionally robust chance-constrained optimization with Wasserstein metric
- Data-driven stochastic programming with distributionally robust constraints under Wasserstein distance: asymptotic properties
- Decomposition methods for Wasserstein-based data-driven distributionally robust problems
- Data-driven risk-averse stochastic optimization with Wasserstein metric
- Data-driven nonparametric robust control under dependence uncertainty
- Distributionally robust optimal control and MDP modeling
- Distributionally Robust Control of Constrained Stochastic Systems
- Distributionally robust modeling of optimal control
Cited in
(23)- Data-driven distributionally robust optimization using the Wasserstein metric: performance guarantees and tractable reformulations
- Data-driven distributionally robust chance-constrained optimization with Wasserstein metric
- Risk-averse autonomous systems: a brief history and recent developments from the perspective of optimal control
- Finite-time convergent zeroing neural network for solving time-varying algebraic Riccati equations
- Frameworks and results in distributionally robust optimization
- Minimax Q-learning control for linear systems using the Wasserstein metric
- GPI-based design for partially unknown nonlinear two-player zero-sum games
- Robust Markov Decision Processes with Data-Driven, Distance-Based Ambiguity Sets
- Distributional Robustness in Minimax Linear Quadratic Control with Wasserstein Distance
- Safe, learning-based MPC for highway driving under Lane-change uncertainty: a distributionally robust approach
- Chance-constrained optimization under limited distributional information: a review of reformulations based on sampling and distributional robustness
- Data‐driven predictive control for a class of uncertain control‐affine systems
- Bayesian Stochastic Gradient Descent for Stochastic Optimization with Streaming Input Data
- Robust Q-learning algorithm for Markov decision processes under Wasserstein uncertainty
- Sequential decision-making under uncertainty: a robust MDPs review
- Optimal transport-based distributionally robust optimization with polynomial uncertainty
- Distributionally robust LQG control under distributed uncertainty
- Distributionally robust optimization
- A robust optimal control problem with moment constraints on distribution: theoretical analysis and an algorithm
- Adaptive dynamic programming and distributionally robust optimal control of linear stochastic system using the Wasserstein metric
- Distributionally robust system identification for continuous fermentation nonlinear switched system under moment uncertainty of experimental data
- Anti-cancer drug administration in cancer treatment under stochastic disturbances: modeling and numerical optimization algorithms
- A decomposition algorithm for distributionally robust chance-constrained programs with polyhedral ambiguity set
This page was built for publication: Wasserstein Distributionally Robust Stochastic Control: A Data-Driven Approach
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4957639)
