Deep Dictionary-Free Method for Identifying Linear Model of Nonlinear System with Input Delay

Patrik Valábek, Marek Wadinger, Michal Kvasnica, Martin Klaučo † · 2025

Read Deep Dictionary-Free Method for Identifying Linear Model of Nonlinear System with Input Delay by Patrik Valábek, Marek Wadinger, Michal Kvasnica, Martin Klaučo † (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2511-04451.txt.

Description

Patrik Valábek and three colleagues model nonlinear systems whose inputs arrive late. Delays make prediction and control hard, since the present state depends on the past. Their approach lifts the system into a space where its motion looks linear, using the Koopman operator.

A deep network with memory layers learns that lifted form on its own. The memory layers store past inputs, so no fixed dictionary of basis functions is needed. Older dictionary methods must guess those functions in advance, which often limits accuracy. A learned, delay-aware linear model could make control design for such systems far simpler.

Catalog record

CREATORPatrik Valábek, Marek Wadinger, Michal Kvasnica, Martin Klaučo †
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE25 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2511-04451

Browse the full library · About the Tower of Babel