Nonparametric Sparse Online Learning of the Koopman Operator

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Hou, Boya, Sanjari, Sina, Dahlin, Nathan, Koppel, Alec, Bose, Subhonmesh
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917408787136512
author Hou, Boya
Sanjari, Sina
Dahlin, Nathan
Koppel, Alec
Bose, Subhonmesh
author_facet Hou, Boya
Sanjari, Sina
Dahlin, Nathan
Koppel, Alec
Bose, Subhonmesh
contents The Koopman operator provides a powerful framework for representing the dynamics of general nonlinear dynamical systems. However, existing data-driven approaches to learning the Koopman operator rely on batch data. In this work, we present a sparse online learning algorithm that learns the Koopman operator iteratively via stochastic approximation, with explicit control over model complexity and provable convergence guarantees. Specifically, we study the Koopman operator via its action on the reproducing kernel Hilbert space (RKHS), and address the mis-specified scenario where the dynamics may escape the chosen RKHS. In this mis-specified setting, we relate the Koopman operator to the conditional mean embeddings (CME) operator. We further establish both asymptotic and finite-time convergence guarantees for our learning algorithm in mis-specified setting, with trajectory-based sampling where the data arrive sequentially over time. Numerical experiments demonstrate the algorithm's capability to learn unknown nonlinear dynamics.
format Preprint
id arxiv_https___arxiv_org_abs_2405_07432
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Nonparametric Sparse Online Learning of the Koopman Operator
Hou, Boya
Sanjari, Sina
Dahlin, Nathan
Koppel, Alec
Bose, Subhonmesh
Machine Learning
Systems and Control
46E22, 47B32, 93B40, 93E12, 93E35
The Koopman operator provides a powerful framework for representing the dynamics of general nonlinear dynamical systems. However, existing data-driven approaches to learning the Koopman operator rely on batch data. In this work, we present a sparse online learning algorithm that learns the Koopman operator iteratively via stochastic approximation, with explicit control over model complexity and provable convergence guarantees. Specifically, we study the Koopman operator via its action on the reproducing kernel Hilbert space (RKHS), and address the mis-specified scenario where the dynamics may escape the chosen RKHS. In this mis-specified setting, we relate the Koopman operator to the conditional mean embeddings (CME) operator. We further establish both asymptotic and finite-time convergence guarantees for our learning algorithm in mis-specified setting, with trajectory-based sampling where the data arrive sequentially over time. Numerical experiments demonstrate the algorithm's capability to learn unknown nonlinear dynamics.
title Nonparametric Sparse Online Learning of the Koopman Operator
topic Machine Learning
Systems and Control
46E22, 47B32, 93B40, 93E12, 93E35
url https://arxiv.org/abs/2405.07432