Arm order recognition in multi-armed bandit problem with laser chaos time series

Abstract By exploiting ultrafast and irregular time series generated by lasers with delayed feedback, we have previously demonstrated a scalable algorithm to solve multi-armed bandit (MAB) problems utilizing the time-division multiplexing of laser chaos time series. Although the algorithm detects th...

Descripción completa

Guardado en:
Detalles Bibliográficos
Autores principales: Naoki Narisawa, Nicolas Chauvet, Mikio Hasegawa, Makoto Naruse
Formato: article
Lenguaje:EN
Publicado: Nature Portfolio 2021
Materias:
R
Q
Acceso en línea:https://doaj.org/article/82e24ef171b342e0baa70ab605b865cc
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
id oai:doaj.org-article:82e24ef171b342e0baa70ab605b865cc
record_format dspace
spelling oai:doaj.org-article:82e24ef171b342e0baa70ab605b865cc2021-12-02T13:20:04ZArm order recognition in multi-armed bandit problem with laser chaos time series10.1038/s41598-021-83726-82045-2322https://doaj.org/article/82e24ef171b342e0baa70ab605b865cc2021-02-01T00:00:00Zhttps://doi.org/10.1038/s41598-021-83726-8https://doaj.org/toc/2045-2322Abstract By exploiting ultrafast and irregular time series generated by lasers with delayed feedback, we have previously demonstrated a scalable algorithm to solve multi-armed bandit (MAB) problems utilizing the time-division multiplexing of laser chaos time series. Although the algorithm detects the arm with the highest reward expectation, the correct recognition of the order of arms in terms of reward expectations is not achievable. Here, we present an algorithm where the degree of exploration is adaptively controlled based on confidence intervals that represent the estimation accuracy of reward expectations. We have demonstrated numerically that our approach did improve arm order recognition accuracy significantly, along with reduced dependence on reward environments, and the total reward is almost maintained compared with conventional MAB methods. This study applies to sectors where the order information is critical, such as efficient allocation of resources in information and communications technology.Naoki NarisawaNicolas ChauvetMikio HasegawaMakoto NaruseNature PortfolioarticleMedicineRScienceQENScientific Reports, Vol 11, Iss 1, Pp 1-11 (2021)
institution DOAJ
collection DOAJ
language EN
topic Medicine
R
Science
Q
spellingShingle Medicine
R
Science
Q
Naoki Narisawa
Nicolas Chauvet
Mikio Hasegawa
Makoto Naruse
Arm order recognition in multi-armed bandit problem with laser chaos time series
description Abstract By exploiting ultrafast and irregular time series generated by lasers with delayed feedback, we have previously demonstrated a scalable algorithm to solve multi-armed bandit (MAB) problems utilizing the time-division multiplexing of laser chaos time series. Although the algorithm detects the arm with the highest reward expectation, the correct recognition of the order of arms in terms of reward expectations is not achievable. Here, we present an algorithm where the degree of exploration is adaptively controlled based on confidence intervals that represent the estimation accuracy of reward expectations. We have demonstrated numerically that our approach did improve arm order recognition accuracy significantly, along with reduced dependence on reward environments, and the total reward is almost maintained compared with conventional MAB methods. This study applies to sectors where the order information is critical, such as efficient allocation of resources in information and communications technology.
format article
author Naoki Narisawa
Nicolas Chauvet
Mikio Hasegawa
Makoto Naruse
author_facet Naoki Narisawa
Nicolas Chauvet
Mikio Hasegawa
Makoto Naruse
author_sort Naoki Narisawa
title Arm order recognition in multi-armed bandit problem with laser chaos time series
title_short Arm order recognition in multi-armed bandit problem with laser chaos time series
title_full Arm order recognition in multi-armed bandit problem with laser chaos time series
title_fullStr Arm order recognition in multi-armed bandit problem with laser chaos time series
title_full_unstemmed Arm order recognition in multi-armed bandit problem with laser chaos time series
title_sort arm order recognition in multi-armed bandit problem with laser chaos time series
publisher Nature Portfolio
publishDate 2021
url https://doaj.org/article/82e24ef171b342e0baa70ab605b865cc
work_keys_str_mv AT naokinarisawa armorderrecognitioninmultiarmedbanditproblemwithlaserchaostimeseries
AT nicolaschauvet armorderrecognitioninmultiarmedbanditproblemwithlaserchaostimeseries
AT mikiohasegawa armorderrecognitioninmultiarmedbanditproblemwithlaserchaostimeseries
AT makotonaruse armorderrecognitioninmultiarmedbanditproblemwithlaserchaostimeseries
_version_ 1718393221135990784