Reservoir Computing in materio with LEDs

Matthew Nicholas Dale; Julian Francis Miller; Susan Stepney; Martin Albrecht Trefzer

Reservoir Computing in materio with LEDs

Matthew Nicholas Dale, Julian Francis Miller, Susan Stepney, Martin Albrecht Trefzer

Research output: Contribution to conference › Paper › peer-review

Abstract

We have shown that the Reservoir Computing framework transfers to complex substrates, and that performance can increase significantly when we control and manipulate input-output mappings and external perturbation through computer-controlled evolution. We have implemented a new example of the hardware-based reservoir methodology. We have two new types of reservoirs based on Light Emitting Diodes (LEDs) and resistors. Results show that unconstrained computer-controlled evolution can exploit the net effect of variations in components (resistors and diodes) to form a single reservoir competitive to previous findings.

Original language	English
Publication status	Published - 2017
Event	International Conference on Unconventional Computation and Natural Computation - Arkansas, Fayetteville, United States Duration: 5 Jun 2017 → 9 Jun 2017 Conference number: 16 https://ucnc2017.csce.uark.edu

Conference

Conference	International Conference on Unconventional Computation and Natural Computation
Abbreviated title	UCNC
Country/Territory	United States
City	Fayetteville
Period	5/06/17 → 9/06/17
Internet address	https://ucnc2017.csce.uark.edu

Keywords

evolution in materio
Unconventional computing
evolvable hardware
natural computation
Carbon Nanotubes (CNTs)
led diode

Access to Document

https://www-users.cs.york.ac.uk/susan/bib/ss/nonstd/ucnc2017-leds.pdf

Cite this

@conference{ee0851d5c11a4b03bfbbb2e475a8e800,

title = "Reservoir Computing in materio with LEDs",

abstract = "We have shown that the Reservoir Computing framework transfers to complex substrates, and that performance can increase significantly when we control and manipulate input-output mappings and external perturbation through computer-controlled evolution. We have implemented a new example of the hardware-based reservoir methodology. We have two new types of reservoirs based on Light Emitting Diodes (LEDs) and resistors. Results show that unconstrained computer-controlled evolution can exploit the net effect of variations in components (resistors and diodes) to form a single reservoir competitive to previous findings.",

keywords = "evolution in materio, Unconventional computing, evolvable hardware, natural computation, Carbon Nanotubes (CNTs), led diode",

author = "Dale, {Matthew Nicholas} and Miller, {Julian Francis} and Susan Stepney and Trefzer, {Martin Albrecht}",

year = "2017",

language = "English",

note = "International Conference on Unconventional Computation and Natural Computation, UCNC ; Conference date: 05-06-2017 Through 09-06-2017",

url = "https://ucnc2017.csce.uark.edu",

}

TY - CONF

T1 - Reservoir Computing in materio with LEDs

AU - Dale, Matthew Nicholas

AU - Miller, Julian Francis

AU - Stepney, Susan

AU - Trefzer, Martin Albrecht

N1 - Conference code: 16

PY - 2017

Y1 - 2017

N2 - We have shown that the Reservoir Computing framework transfers to complex substrates, and that performance can increase significantly when we control and manipulate input-output mappings and external perturbation through computer-controlled evolution. We have implemented a new example of the hardware-based reservoir methodology. We have two new types of reservoirs based on Light Emitting Diodes (LEDs) and resistors. Results show that unconstrained computer-controlled evolution can exploit the net effect of variations in components (resistors and diodes) to form a single reservoir competitive to previous findings.

AB - We have shown that the Reservoir Computing framework transfers to complex substrates, and that performance can increase significantly when we control and manipulate input-output mappings and external perturbation through computer-controlled evolution. We have implemented a new example of the hardware-based reservoir methodology. We have two new types of reservoirs based on Light Emitting Diodes (LEDs) and resistors. Results show that unconstrained computer-controlled evolution can exploit the net effect of variations in components (resistors and diodes) to form a single reservoir competitive to previous findings.

KW - evolution in materio

KW - Unconventional computing

KW - evolvable hardware

KW - natural computation

KW - Carbon Nanotubes (CNTs)

KW - led diode

M3 - Paper

T2 - International Conference on Unconventional Computation and Natural Computation

Y2 - 5 June 2017 through 9 June 2017

ER -