Photonic spiking neural network with resonant tunnelling diode optoelectronic neurons

Matěj Hejda; Juan Alanis; Ignacio Ortega-Piwonka; José Figueiredo; João Lourenco; Julien Javaloyes; Bruno Romeira; Antonio Hurtado

Photonic spiking neural network with resonant tunnelling diode optoelectronic neurons

Matěj Hejda, Juan Alanis, Ignacio Ortega-Piwonka, José Figueiredo, João Lourenco, Julien Javaloyes, Bruno Romeira, Antonio Hurtado

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution book

26 Downloads (Pure)

Abstract

We report high-speed, energy-efficient artificial optoelectronic spiking neurons based upon resonant tunnelling diodes (RTD). Using these, we demonstrate a photonic spiking neural network (perceptron) numerical model for classification of spatiotemporal pulse patterns with 94% accuracy.

Original language	English
Title of host publication	Conference on Lasers and Electro-Optics
Place of Publication	San José, CA.
Publication status	Published - 18 May 2022

Keywords

optoelectric neurons
Artificial Intelligence (AI)
neural networks
semiconductor devices
resonant tunnelling diodes (RTDs)
double-barrier quantum well (DBQW)

Access to Document

Hejda-etal-CLEO-2022-Photonic-spiking-neural-network-with-resonant-tunnelling-diode-optoelectronic-neuronsAccepted author manuscript, 1.7 MBLicence: Strath_1

Cite this

@inproceedings{0798e8386d4f4692bfaadb1421a0aaa6,

title = "Photonic spiking neural network with resonant tunnelling diode optoelectronic neurons",

abstract = "We report high-speed, energy-efficient artificial optoelectronic spiking neurons based upon resonant tunnelling diodes (RTD). Using these, we demonstrate a photonic spiking neural network (perceptron) numerical model for classification of spatiotemporal pulse patterns with 94\% accuracy.",

keywords = "optoelectric neurons, Artificial Intelligence (AI), neural networks, semiconductor devices, resonant tunnelling diodes (RTDs), double-barrier quantum well (DBQW)",

author = "Mat{\v e}j Hejda and Juan Alanis and Ignacio Ortega-Piwonka and Jos{\'e} Figueiredo and Jo{\~a}o Lourenco and Julien Javaloyes and Bruno Romeira and Antonio Hurtado",

year = "2022",

month = may,

day = "18",

language = "English",

booktitle = "Conference on Lasers and Electro-Optics",

}

TY - GEN

T1 - Photonic spiking neural network with resonant tunnelling diode optoelectronic neurons

AU - Hejda, Matěj

AU - Alanis, Juan

AU - Ortega-Piwonka, Ignacio

AU - Figueiredo, José

AU - Lourenco, João

AU - Javaloyes, Julien

AU - Romeira, Bruno

AU - Hurtado, Antonio

PY - 2022/5/18

Y1 - 2022/5/18

N2 - We report high-speed, energy-efficient artificial optoelectronic spiking neurons based upon resonant tunnelling diodes (RTD). Using these, we demonstrate a photonic spiking neural network (perceptron) numerical model for classification of spatiotemporal pulse patterns with 94% accuracy.

AB - We report high-speed, energy-efficient artificial optoelectronic spiking neurons based upon resonant tunnelling diodes (RTD). Using these, we demonstrate a photonic spiking neural network (perceptron) numerical model for classification of spatiotemporal pulse patterns with 94% accuracy.

KW - optoelectric neurons

KW - Artificial Intelligence (AI)

KW - neural networks

KW - semiconductor devices

KW - resonant tunnelling diodes (RTDs)

KW - double-barrier quantum well (DBQW)

UR - https://opg.optica.org/abstract.cfm?uri=CLEO_SI-2022-SW5E.1

M3 - Conference contribution book

BT - Conference on Lasers and Electro-Optics

CY - San José, CA.

ER -

Photonic spiking neural network with resonant tunnelling diode optoelectronic neurons

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Energy-efficient and high-bandwidth neuromorphic nanophotonic chips for artificial intelligence systems (ChipAI) H2020-FETOPEN

Resonant tunneling diode nano-optoelectronic excitable nodes for neuromorphic spike-based information processing

Cite this

Photonic spiking neural network with resonant tunnelling diode optoelectronic neurons

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Projects

Energy-efficient and high-bandwidth neuromorphic nanophotonic chips for artificial intelligence systems (ChipAI) H2020-FETOPEN

Research output

Resonant tunneling diode nano-optoelectronic excitable nodes for neuromorphic spike-based information processing

Cite this