Dale, Matthew, Miller, Julian F orcid.org/0000-0002-7692-9655, Stepney, Susan orcid.org/0000-0003-3146-5401 et al. (1 more author) (2019) A substrate-independent framework to characterize reservoir computers. Proceedings of the Royal Society of London. Series A, Mathematical and Physical Sciences. 20180723. ISSN 1364-5021
Abstract
The reservoir computing (RC) framework states that any nonlinear, input-driven dynamical system (the reservoir) exhibiting properties such as a fading memory and input separability can be trained to perform computational tasks. This broad inclusion of systems has led to many new physical substrates for RC. Properties essential for reservoirs to compute are tuned through reconfiguration of the substrate, such as change in virtual topology or physical morphology. As a result, each substrate possesses a unique 'quality'-obtained through reconfiguration-to realize different reservoirs for different tasks. Here we describe an experimental framework to characterize the quality of potentially any substrate for RC. Our framework reveals that a definition of quality is not only useful to compare substrates, but can help map the non-trivial relationship between properties and task performance. In the wider context, the framework offers a greater understanding as to what makes a dynamical system compute, helping improve the design of future substrates for RC.
Metadata
Item Type: | Article |
---|---|
Authors/Creators: |
|
Copyright, Publisher and Additional Information: | Funding Information: Data accessibility. The research material (source code and data) can be accessed at github.com/MaterialMan/ CHARC-Framework.git. Authors’ contributions. M.D. designed the framework and performed experimental work. All authors discussed the methods and results, contributed to the writing of the manuscript, and gave their final approval for publication. Competing interests. We declare we have no competing interests. Funding. This work was part-funded by a Defence Science and Technology Laboratory (DSTL) PhD studentship, and part-funded by the SpInsired project, EPSRC grant no. EP/R032823/1. Publisher Copyright: © 2019 The Authors. Published by the Royal Society under the terms of the Creative Commons Attribution License |
Keywords: | Characterization,Physical computation,Reservoir computing |
Dates: |
|
Institution: | The University of York |
Academic Units: | The University of York > Faculty of Sciences (York) > Computer Science (York) The University of York > Faculty of Sciences (York) > Electronic Engineering (York) |
Depositing User: | Pure (York) |
Date Deposited: | 05 Aug 2019 15:40 |
Last Modified: | 21 Jan 2025 17:41 |
Status: | Published |
Refereed: | Yes |
Related URLs: | |
Open Archives Initiative ID (OAI ID): | oai:eprints.whiterose.ac.uk:149352 |