Finite-data error bounds for Koopman-based prediction and control

DOI10.1007/S00332-022-09862-1arXiv2108.07102WikidataQ122930249 ScholiaQ122930249MaRDI QIDQ6375358FDOQ6375358

Authors: Feliks Nüske, Sebastian Peitz, Friedrich Philipp, M. Schaller, Karl Worthmann 

Publication date: 16 August 2021

Abstract: The Koopman operator has become an essential tool for data-driven approximation of dynamical (control) systems, e.g., via extended dynamic mode decomposition. Despite its popularity, convergence results and, in particular, error bounds are still scarce. In this paper, we derive probabilistic bounds for the approximation error and the prediction error depending on the number of training data points; for both ordinary and stochastic differential equations while using either ergodic trajectories or i.i.d. samples. We illustrate these bounds by means of an example with the Ornstein-Uhlenbeck process. Moreover, we extend our analysis to (stochastic) nonlinear control-affine systems. We prove error estimates for a previously proposed approach that exploits the linearity of the Koopman generator to obtain a bilinear surrogate control system and, thus, circumvents the curse of dimensionality since the system is not autonomized by augmenting the state by the control inputs. To the best of our knowledge, this is the first finite-data error analysis in the stochastic and/or control setting. Finally, we demonstrate the effectiveness of the bilinear approach by comparing it with state-of-the-art techniques showing its superiority whenever state and control are coupled.

Mathematics Subject Classification ID

Dynamical systems and their relations with probability theory and stochastic processes (37A50) Functional analytic techniques in dynamical systems; zeta functions, (Ruelle-Frobenius) transfer operators, etc. (37C30) System identification (93B30) Approximation methods and numerical treatment of dynamical systems (37M99) Groups and semigroups of linear operators (47D03) One-parameter semigroups and linear evolution equations (47D06) Operator-theoretic methods (93B28)

This page was built for publication: Finite-data error bounds for Koopman-based prediction and control

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6375358)