Restricted Reservoirs on Heterogeneous Timescales

Chester Wringe; Susan Stepney; Martin Albrecht Trefzer

doi:10.1007/978-3-031-72359-9_13

Restricted Reservoirs on Heterogeneous Timescales

Chester Wringe, Susan Stepney, Martin Albrecht Trefzer

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

The Reservoir Computing model is one that is suited to computing with physical materials. However, the limitations of those materials can lead to low computational power. To address this, we plan to combine reservoirs operating on different timescales together to create a heterogeneous reservoir. We simulate this using a new multiple timescale ESN model. We also introduce “mock materials” so that future works may focus on combining different materials to study the effect. We benchmark our multi-timescale ESN on a multiple timescale problem, MSO.

Original language	English
Title of host publication	Lecture Notes in Computer Science
Subtitle of host publication	Artificial Neural Networks and Machine Learning - ICANN
Pages	168-183
ISBN (Electronic)	978-3-031-72359-9
DOIs	https://doi.org/10.1007/978-3-031-72359-9_13
Publication status	Published - 18 Sept 2024

Publication series

Name	Lecture Notes in Computer Science
Volume	15025
ISSN (Print)	0302-9743

Bibliographical note

This is an author-produced version of the published paper. Uploaded in accordance with the University’s Research Publications and Open Access policy.

Keywords

reservoir computing
heterogeneous timescales

Access to Document

10.1007/978-3-031-72359-9_13

_ICANN_2024__Restricted_Reservoirs_on_Heterogeneous_Timescales
This is an author-produced version of the published paper. Uploaded in accordance with the University’s Research Publications and Open Access policy.
Accepted author manuscript, 563 KBLicence: CC BY

Cite this

@inproceedings{6509c3dc7d6a493d9b62be6e0a6d15b0,

title = "Restricted Reservoirs on Heterogeneous Timescales",

abstract = "The Reservoir Computing model is one that is suited to computing with physical materials. However, the limitations of those materials can lead to low computational power. To address this, we plan to combine reservoirs operating on different timescales together to create a heterogeneous reservoir. We simulate this using a new multiple timescale ESN model. We also introduce “mock materials” so that future works may focus on combining different materials to study the effect. We benchmark our multi-timescale ESN on a multiple timescale problem, MSO.",

keywords = "reservoir computing, heterogeneous timescales",

author = "Chester Wringe and Susan Stepney and Trefzer, {Martin Albrecht}",

note = "This is an author-produced version of the published paper. Uploaded in accordance with the University{\textquoteright}s Research Publications and Open Access policy.",

year = "2024",

month = sep,

day = "18",

doi = "10.1007/978-3-031-72359-9_13",

language = "English",

isbn = "978-3-031-72358-2",

series = "Lecture Notes in Computer Science",

pages = "168--183",

booktitle = "Lecture Notes in Computer Science",

}

TY - GEN

T1 - Restricted Reservoirs on Heterogeneous Timescales

AU - Wringe, Chester

AU - Stepney, Susan

AU - Trefzer, Martin Albrecht

N1 - This is an author-produced version of the published paper. Uploaded in accordance with the University’s Research Publications and Open Access policy.

PY - 2024/9/18

Y1 - 2024/9/18

N2 - The Reservoir Computing model is one that is suited to computing with physical materials. However, the limitations of those materials can lead to low computational power. To address this, we plan to combine reservoirs operating on different timescales together to create a heterogeneous reservoir. We simulate this using a new multiple timescale ESN model. We also introduce “mock materials” so that future works may focus on combining different materials to study the effect. We benchmark our multi-timescale ESN on a multiple timescale problem, MSO.

AB - The Reservoir Computing model is one that is suited to computing with physical materials. However, the limitations of those materials can lead to low computational power. To address this, we plan to combine reservoirs operating on different timescales together to create a heterogeneous reservoir. We simulate this using a new multiple timescale ESN model. We also introduce “mock materials” so that future works may focus on combining different materials to study the effect. We benchmark our multi-timescale ESN on a multiple timescale problem, MSO.

KW - reservoir computing

KW - heterogeneous timescales

U2 - 10.1007/978-3-031-72359-9_13

DO - 10.1007/978-3-031-72359-9_13

M3 - Conference contribution

SN - 978-3-031-72358-2

T3 - Lecture Notes in Computer Science

SP - 168

EP - 183

BT - Lecture Notes in Computer Science

ER -