Model-size reduction for reservoir computing by concatenating internal states through time

Abstract Reservoir computing (RC) is a machine learning algorithm that can learn complex time series from data very rapidly based on the use of high-dimensional dynamical systems, such as random networks of neurons, called “reservoirs.” To implement RC in edge computing, it is highly important to re...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Yusuke Sakemi, Kai Morino, Timothée Leleu, Kazuyuki Aihara
Formato:	article
Lenguaje:	EN
Publicado:	Nature Portfolio 2020
Materias:	Medicine R Science Q
Acceso en línea:	https://doaj.org/article/739fe033c48f4169b1048236ff7c25e0
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:739fe033c48f4169b1048236ff7c25e0
record_format	dspace
spelling	oai:doaj.org-article:739fe033c48f4169b1048236ff7c25e02021-12-02T11:43:51ZModel-size reduction for reservoir computing by concatenating internal states through time10.1038/s41598-020-78725-02045-2322https://doaj.org/article/739fe033c48f4169b1048236ff7c25e02020-12-01T00:00:00Zhttps://doi.org/10.1038/s41598-020-78725-0https://doaj.org/toc/2045-2322Abstract Reservoir computing (RC) is a machine learning algorithm that can learn complex time series from data very rapidly based on the use of high-dimensional dynamical systems, such as random networks of neurons, called “reservoirs.” To implement RC in edge computing, it is highly important to reduce the amount of computational resources that RC requires. In this study, we propose methods that reduce the size of the reservoir by inputting the past or drifting states of the reservoir to the output layer at the current time step. To elucidate the mechanism of model-size reduction, the proposed methods are analyzed based on information processing capacity proposed by Dambre et al. (Sci Rep 2:514, 2012). In addition, we evaluate the effectiveness of the proposed methods on time-series prediction tasks: the generalized Hénon-map and NARMA. On these tasks, we found that the proposed methods were able to reduce the size of the reservoir up to one tenth without a substantial increase in regression error.Yusuke SakemiKai MorinoTimothée LeleuKazuyuki AiharaNature PortfolioarticleMedicineRScienceQENScientific Reports, Vol 10, Iss 1, Pp 1-13 (2020)
institution	DOAJ
collection	DOAJ
language	EN
topic	Medicine R Science Q
spellingShingle	Medicine R Science Q Yusuke Sakemi Kai Morino Timothée Leleu Kazuyuki Aihara Model-size reduction for reservoir computing by concatenating internal states through time
description	Abstract Reservoir computing (RC) is a machine learning algorithm that can learn complex time series from data very rapidly based on the use of high-dimensional dynamical systems, such as random networks of neurons, called “reservoirs.” To implement RC in edge computing, it is highly important to reduce the amount of computational resources that RC requires. In this study, we propose methods that reduce the size of the reservoir by inputting the past or drifting states of the reservoir to the output layer at the current time step. To elucidate the mechanism of model-size reduction, the proposed methods are analyzed based on information processing capacity proposed by Dambre et al. (Sci Rep 2:514, 2012). In addition, we evaluate the effectiveness of the proposed methods on time-series prediction tasks: the generalized Hénon-map and NARMA. On these tasks, we found that the proposed methods were able to reduce the size of the reservoir up to one tenth without a substantial increase in regression error.
format	article
author	Yusuke Sakemi Kai Morino Timothée Leleu Kazuyuki Aihara
author_facet	Yusuke Sakemi Kai Morino Timothée Leleu Kazuyuki Aihara
author_sort	Yusuke Sakemi
title	Model-size reduction for reservoir computing by concatenating internal states through time
title_short	Model-size reduction for reservoir computing by concatenating internal states through time
title_full	Model-size reduction for reservoir computing by concatenating internal states through time
title_fullStr	Model-size reduction for reservoir computing by concatenating internal states through time
title_full_unstemmed	Model-size reduction for reservoir computing by concatenating internal states through time
title_sort	model-size reduction for reservoir computing by concatenating internal states through time
publisher	Nature Portfolio
publishDate	2020
url	https://doaj.org/article/739fe033c48f4169b1048236ff7c25e0
work_keys_str_mv	AT yusukesakemi modelsizereductionforreservoircomputingbyconcatenatinginternalstatesthroughtime AT kaimorino modelsizereductionforreservoircomputingbyconcatenatinginternalstatesthroughtime AT timotheeleleu modelsizereductionforreservoircomputingbyconcatenatinginternalstatesthroughtime AT kazuyukiaihara modelsizereductionforreservoircomputingbyconcatenatinginternalstatesthroughtime
_version_	1718395348844544000

Model-size reduction for reservoir computing by concatenating internal states through time

Ejemplares similares