Dictionary learning in Extended Dynamic Mode Decomposition using a reservoir computer

Research output: Contribution in Book/Catalog/Report/Conference proceedingConference contribution

56 Downloads (Pure)

Abstract

We aim at improving extended dynamic mode decomposition that allows to linearize nonlinear systems.
Indeed, the EDMD algorithm provides a finite-dimensional representation of the Koopman operator.
Finally, the reservoir computer is trained to produce an efficient dictionary
Original languageEnglish
Title of host publicationBook of Abstracts
Subtitle of host publication38th Benelux Meeting on Systems and Control
EditorsPanos Patrinos, Ivan Markovsky, Wim Michiels, Jan Swevers, Steffen Waldherr
PublisherKU Leuven
Pages42
Number of pages211
ISBN (Print)9789082825954
Publication statusPublished - 19 Mar 2019

Fingerprint

Dive into the research topics of 'Dictionary learning in Extended Dynamic Mode Decomposition using a reservoir computer'. Together they form a unique fingerprint.

Cite this