A Redox‐Based Ion‐Gating Reservoir, Utilizing Double Reservoir States in Drain and Gate Nonlinear Responses

Herein, physical reservoir computing with a redox‐based ion‐gating reservoir (redox‐IGR) comprising Li x WO 3 thin film and lithium‐ion conducting glass ceramic (LICGC) is demonstrated. The subject redox‐IGR successfully solves a second‐order nonlinear dynamic equation by utilizing voltage pulse dri...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:Advanced intelligent systems Ročník 5; číslo 9
Hlavní autoři: Wada, Tomoki, Nishioka, Daiki, Namiki, Wataru, Tsuchiya, Takashi, Higuchi, Tohru, Terabe, Kazuya
Médium: Journal Article
Jazyk:angličtina
Vydáno: Weinheim John Wiley & Sons, Inc 01.09.2023
Wiley
Témata:
all-solid-state
Autoregressive moving average
Computation
Electrodes
Electrolytes
Errors
Glass ceramics
Integrated circuits
ion-gating reservoir
ion-gating transistor
lithium ion
Lithium ions
Memory tasks
nanoionics
neuromorphic
Nonlinear dynamics
Nonlinear response
Pattern recognition
Thin films
Transistors
ISSN:2640-4567, 2640-4567
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:Herein, physical reservoir computing with a redox‐based ion‐gating reservoir (redox‐IGR) comprising Li x WO 3 thin film and lithium‐ion conducting glass ceramic (LICGC) is demonstrated. The subject redox‐IGR successfully solves a second‐order nonlinear dynamic equation by utilizing voltage pulse driven ion‐gating in a Li x WO 3 channel to enable reservoir computing. Under the normal conditions, in which only the drain current ( I D ) is used for the reservoir states, the lowest prediction error is 8.15 × 10 −4 . Performance is enhanced by the addition of I G to the reservoir states, resulting in a significant lowering of the prediction error to 5.39 × 10 −4 , which is noticeably lower than other types of physical reservoirs (memristors and spin torque oscillators) reported to date. A second‐order nonlinear autoregressive moving average (NARMA2) task, a typical benchmark of reservoir computing, is also performed with the IGR and good performance is achieved, with a normalized mean square error (NMSE) of 0.163. A short‐term memory task is performed to investigate an enhancement mechanism resulting from the I G addition. An increase in memory capacity, from 2.35 without I G to 3.57 with I G , is observed in the forgetting curves, indicating that enhancement of both high dimensionality and memory capacity is attributed to the origin of the performance improvement.
Bibliografie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:2640-4567
2640-4567
DOI:10.1002/aisy.202300123