Please use this identifier to cite or link to this item: http://dx.doi.org/10.14279/depositonce-12760
For citation please use:
Main Title: Emulating complex networks with a single delay differential equation
Author(s): Stelzer, Florian
Yanchuk, Serhiy
Type: Article
URI: https://depositonce.tu-berlin.de/handle/11303/13987
http://dx.doi.org/10.14279/depositonce-12760
License: https://creativecommons.org/licenses/by/4.0/
Abstract: A single dynamical system with time-delayed feedback can emulate networks. This property of delay systems made them extremely useful tools for Machine-Learning applications. Here, we describe several possible setups, which allow emulating multilayer (deep) feed-forward networks as well as recurrent networks of coupled discrete maps with arbitrary adjacency matrix by a single system with delayed feedback. While the network’s size can be arbitrary, the generating delay system can have a low number of variables, including a scalar case.
Subject(s): multilayer
network
neural network
Issue Date: 6-Jun-2021
Date Available: 6-Dec-2021
Language Code: en
DDC Class: 510 Mathematik
Sponsor/Funder: DFG, 411803875, Dynamik gekoppelter Systeme mit Zeitverzögerungen und deren Anwendungen
DFG, 183049896, GRK 1740: Dynamische Phänomene in komplexen Netzwerken: Grundlagen und Anwendungen
TU Berlin, Open-Access-Mittel – 2021
Journal Title: The European Physical Journal Special Topics
Publisher: Springer Nature
Volume: 230
Issue: 14-15
Publisher DOI: 10.1140/epjs/s11734-021-00162-5
Page Start: 2865
Page End: 2874
EISSN: 1951-6401
ISSN: 1951-6355
TU Affiliation(s): Fak. 2 Mathematik und Naturwissenschaften » Inst. Mathematik » FG Numerische Mathematik
Appears in Collections:Technische Universität Berlin » Publications

Files in This Item:
stelzer_yanchuk_2021.pdf
Format: Adobe PDF | Size: 672.45 kB
DownloadShow Preview
Thumbnail

Item Export Bar

This item is licensed under a Creative Commons License Creative Commons