Post-processing methods for delay embedding and feature scaling of reservoir computers
Abstract Reservoir computing is a machine learning method that is well-suited for complex time series prediction tasks. Both delay embedding and the projection of input data into a higher-dimensional space play important roles in enabling accurate predictions. We establish simple post-processing met...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2025-01-01
|
| Series: | Communications Engineering |
| Online Access: | https://doi.org/10.1038/s44172-024-00330-0 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832571630588526592 |
|---|---|
| author | Jonnel Jaurigue Joshua Robertson Antonio Hurtado Lina Jaurigue Kathy Lüdge |
| author_facet | Jonnel Jaurigue Joshua Robertson Antonio Hurtado Lina Jaurigue Kathy Lüdge |
| author_sort | Jonnel Jaurigue |
| collection | DOAJ |
| description | Abstract Reservoir computing is a machine learning method that is well-suited for complex time series prediction tasks. Both delay embedding and the projection of input data into a higher-dimensional space play important roles in enabling accurate predictions. We establish simple post-processing methods that train on past node states at uniformly or randomly-delayed timeshifts. These methods improve reservoir computer prediction performance through increased feature dimension and/or better delay embedding. Here we introduce the multi-random-timeshifting method that randomly recalls previous states of reservoir nodes. The use of multi-random-timeshifting allows for smaller reservoirs while maintaining large feature dimensions, is computationally cheap to optimise, and is our preferred post-processing method. For experimentalists, all our post-processing methods can be translated to readout data sampled from physical reservoirs, which we demonstrate using readout data from an experimentally-realised laser reservoir system. |
| format | Article |
| id | doaj-art-05d36360dce74ea1bc8ce2c9e3579ff3 |
| institution | Kabale University |
| issn | 2731-3395 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Communications Engineering |
| spelling | doaj-art-05d36360dce74ea1bc8ce2c9e3579ff32025-02-02T12:27:01ZengNature PortfolioCommunications Engineering2731-33952025-01-014111310.1038/s44172-024-00330-0Post-processing methods for delay embedding and feature scaling of reservoir computersJonnel Jaurigue0Joshua Robertson1Antonio Hurtado2Lina Jaurigue3Kathy Lüdge4Institut für Physik, Technische Universität IlmenauInstitute of Photonics, SUPA Department of Physics, University of StrathclydeInstitute of Photonics, SUPA Department of Physics, University of StrathclydeInstitut für Physik, Technische Universität IlmenauInstitut für Physik, Technische Universität IlmenauAbstract Reservoir computing is a machine learning method that is well-suited for complex time series prediction tasks. Both delay embedding and the projection of input data into a higher-dimensional space play important roles in enabling accurate predictions. We establish simple post-processing methods that train on past node states at uniformly or randomly-delayed timeshifts. These methods improve reservoir computer prediction performance through increased feature dimension and/or better delay embedding. Here we introduce the multi-random-timeshifting method that randomly recalls previous states of reservoir nodes. The use of multi-random-timeshifting allows for smaller reservoirs while maintaining large feature dimensions, is computationally cheap to optimise, and is our preferred post-processing method. For experimentalists, all our post-processing methods can be translated to readout data sampled from physical reservoirs, which we demonstrate using readout data from an experimentally-realised laser reservoir system.https://doi.org/10.1038/s44172-024-00330-0 |
| spellingShingle | Jonnel Jaurigue Joshua Robertson Antonio Hurtado Lina Jaurigue Kathy Lüdge Post-processing methods for delay embedding and feature scaling of reservoir computers Communications Engineering |
| title | Post-processing methods for delay embedding and feature scaling of reservoir computers |
| title_full | Post-processing methods for delay embedding and feature scaling of reservoir computers |
| title_fullStr | Post-processing methods for delay embedding and feature scaling of reservoir computers |
| title_full_unstemmed | Post-processing methods for delay embedding and feature scaling of reservoir computers |
| title_short | Post-processing methods for delay embedding and feature scaling of reservoir computers |
| title_sort | post processing methods for delay embedding and feature scaling of reservoir computers |
| url | https://doi.org/10.1038/s44172-024-00330-0 |
| work_keys_str_mv | AT jonneljaurigue postprocessingmethodsfordelayembeddingandfeaturescalingofreservoircomputers AT joshuarobertson postprocessingmethodsfordelayembeddingandfeaturescalingofreservoircomputers AT antoniohurtado postprocessingmethodsfordelayembeddingandfeaturescalingofreservoircomputers AT linajaurigue postprocessingmethodsfordelayembeddingandfeaturescalingofreservoircomputers AT kathyludge postprocessingmethodsfordelayembeddingandfeaturescalingofreservoircomputers |