Laplace Transform Based Low-Complexity Learning of Continuous Markov Semigroups

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Kostic, Vladimir R., Lounici, Karim, Halconruy, Hélène, Devergne, Timothée, Novelli, Pietro, Pontil, Massimiliano
Format: Preprint
Veröffentlicht: 2024
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866908395678728192
author Kostic, Vladimir R.
Lounici, Karim
Halconruy, Hélène
Devergne, Timothée
Novelli, Pietro
Pontil, Massimiliano
author_facet Kostic, Vladimir R.
Lounici, Karim
Halconruy, Hélène
Devergne, Timothée
Novelli, Pietro
Pontil, Massimiliano
contents Markov processes serve as a universal model for many real-world random processes. This paper presents a data-driven approach for learning these models through the spectral decomposition of the infinitesimal generator (IG) of the Markov semigroup. The unbounded nature of IGs complicates traditional methods such as vector-valued regression and Hilbert-Schmidt operator analysis. Existing techniques, including physics-informed kernel regression, are computationally expensive and limited in scope, with no recovery guarantees for transfer operator methods when the time-lag is small. We propose a novel method that leverages the IG's resolvent, characterized by the Laplace transform of transfer operators. This approach is robust to time-lag variations, ensuring accurate eigenvalue learning even for small time-lags. Our statistical analysis applies to a broader class of Markov processes than current methods while reducing computational complexity from quadratic to linear in the state dimension. Finally, we illustrate the behaviour of our method in two experiments.
format Preprint
id arxiv_https___arxiv_org_abs_2410_14477
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Laplace Transform Based Low-Complexity Learning of Continuous Markov Semigroups
Kostic, Vladimir R.
Lounici, Karim
Halconruy, Hélène
Devergne, Timothée
Novelli, Pietro
Pontil, Massimiliano
Machine Learning
Statistics Theory
Markov processes serve as a universal model for many real-world random processes. This paper presents a data-driven approach for learning these models through the spectral decomposition of the infinitesimal generator (IG) of the Markov semigroup. The unbounded nature of IGs complicates traditional methods such as vector-valued regression and Hilbert-Schmidt operator analysis. Existing techniques, including physics-informed kernel regression, are computationally expensive and limited in scope, with no recovery guarantees for transfer operator methods when the time-lag is small. We propose a novel method that leverages the IG's resolvent, characterized by the Laplace transform of transfer operators. This approach is robust to time-lag variations, ensuring accurate eigenvalue learning even for small time-lags. Our statistical analysis applies to a broader class of Markov processes than current methods while reducing computational complexity from quadratic to linear in the state dimension. Finally, we illustrate the behaviour of our method in two experiments.
title Laplace Transform Based Low-Complexity Learning of Continuous Markov Semigroups
topic Machine Learning
Statistics Theory
url https://arxiv.org/abs/2410.14477