High speed time series prediction with a photonic spiking neural network built with a single VCSEL

D. Owen-Newns; L. Jaurigue; J. Robertson; A. Adair; J. Jaurigue; K. Lüdge; A. Hurtado

doi:10.1109/ipc60965.2024.10799650

High speed time series prediction with a photonic spiking neural network built with a single VCSEL

D. Owen-Newns, L. Jaurigue, J. Robertson, A. Adair, J. Jaurigue, K. Lüdge, A. Hurtado

Institute Of Photonics

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

Abstract

We report a high speed Neuromorphic Photonic Spiking Neural Network (p-SNN) built with a Vertical Cavity Surface Emitting Laser (VCSEL). Memory is induced in the p-SNN by introducing two separate past points in the time series allowing accurate predictions over a range of future horizons.

Original language	English
Title of host publication	2024 IEEE Photonics Conference (IPC)
Publisher	IEEE
Pages	1-2
Number of pages	2
ISBN (Electronic)	979-8-3503-6195-7
ISBN (Print)	979-8-3503-6196-4
DOIs	https://doi.org/10.1109/ipc60965.2024.10799650
Publication status	Published - 20 Dec 2024
Event	IEEE Photonics Conference 2024 - Rome, Italy Duration: 10 Nov 2024 → 14 Nov 2024 https://ieee-ipc.org/

Publication series

Name	2024 IEEE Photonics Conference (IPC)
Publisher	IEEE
ISSN (Print)	2374-0140
ISSN (Electronic)	2575-274X

Conference

Conference	IEEE Photonics Conference 2024
Abbreviated title	IPC
Country/Territory	Italy
City	Rome
Period	10/11/24 → 14/11/24
Internet address	https://ieee-ipc.org/

Funding

The authors acknowledge support by the UKRI Turing AI Acceleration Fellowships Programme (EP/V025198/1), and by the EU Pathfinder Open project ‘SpikePro’. The authors acknowledge support from the Fraunhofer Centre for Applied Photonics, FCAP.

Keywords

VCSELs
Neuromorphic Photonics
Spiking Neural Networks
Semiconductor Lasers

Access to Document

10.1109/ipc60965.2024.10799650

SPIKING PHOTONIC-ELECTRONIC IC FOR QUICK AND EFFICIENT PROCESSING (SPIKEPro)
Hurtado, A. (Principal Investigator), Clemente, C. (Co-investigator) & Porte Parera, J. (Co-investigator)
European Commission - Horizon Europe + H2020
1/03/24 → 29/02/28
Project: Research
Turing AI Fellowship: PHOTONics for ultrafast Artificial Intelligence
Hurtado, A. (Fellow)
EPSRC (Engineering and Physical Sciences Research Council)
1/01/21 → 31/12/25
Project: Research Fellowship

Cite this

@inproceedings{bf7c718750bb4ce58c6a348ae81c52ea,

title = "High speed time series prediction with a photonic spiking neural network built with a single VCSEL",

abstract = "We report a high speed Neuromorphic Photonic Spiking Neural Network (p-SNN) built with a Vertical Cavity Surface Emitting Laser (VCSEL). Memory is induced in the p-SNN by introducing two separate past points in the time series allowing accurate predictions over a range of future horizons.",

keywords = "VCSELs, Neuromorphic Photonics, Spiking Neural Networks, Semiconductor Lasers",

author = "D. Owen-Newns and L. Jaurigue and J. Robertson and A. Adair and J. Jaurigue and K. L{\"u}dge and A. Hurtado",

year = "2024",

month = dec,

day = "20",

doi = "10.1109/ipc60965.2024.10799650",

language = "English",

isbn = "979-8-3503-6196-4",

series = "2024 IEEE Photonics Conference (IPC)",

publisher = "IEEE",

pages = "1--2",

booktitle = "2024 IEEE Photonics Conference (IPC)",

note = "IEEE Photonics Conference 2024, IPC ; Conference date: 10-11-2024 Through 14-11-2024",

url = "https://ieee-ipc.org/",

}

Owen-Newns, D, Jaurigue, L, Robertson, J, Adair, A, Jaurigue, J, Lüdge, K & Hurtado, A 2024, High speed time series prediction with a photonic spiking neural network built with a single VCSEL. in 2024 IEEE Photonics Conference (IPC). 2024 IEEE Photonics Conference (IPC), IEEE, pp. 1-2, IEEE Photonics Conference 2024, Rome, Italy, 10/11/24. https://doi.org/10.1109/ipc60965.2024.10799650

TY - GEN

T1 - High speed time series prediction with a photonic spiking neural network built with a single VCSEL

AU - Owen-Newns, D.

AU - Jaurigue, L.

AU - Robertson, J.

AU - Adair, A.

AU - Jaurigue, J.

AU - Lüdge, K.

AU - Hurtado, A.

PY - 2024/12/20

Y1 - 2024/12/20

N2 - We report a high speed Neuromorphic Photonic Spiking Neural Network (p-SNN) built with a Vertical Cavity Surface Emitting Laser (VCSEL). Memory is induced in the p-SNN by introducing two separate past points in the time series allowing accurate predictions over a range of future horizons.

AB - We report a high speed Neuromorphic Photonic Spiking Neural Network (p-SNN) built with a Vertical Cavity Surface Emitting Laser (VCSEL). Memory is induced in the p-SNN by introducing two separate past points in the time series allowing accurate predictions over a range of future horizons.

KW - VCSELs

KW - Neuromorphic Photonics

KW - Spiking Neural Networks

KW - Semiconductor Lasers

U2 - 10.1109/ipc60965.2024.10799650

DO - 10.1109/ipc60965.2024.10799650

M3 - Conference contribution book

SN - 979-8-3503-6196-4

T3 - 2024 IEEE Photonics Conference (IPC)

SP - 1

EP - 2

BT - 2024 IEEE Photonics Conference (IPC)

PB - IEEE

T2 - IEEE Photonics Conference 2024

Y2 - 10 November 2024 through 14 November 2024

ER -

High speed time series prediction with a photonic spiking neural network built with a single VCSEL

Abstract

Publication series

Conference

Funding

Keywords

Access to Document

Projects

SPIKING PHOTONIC-ELECTRONIC IC FOR QUICK AND EFFICIENT PROCESSING (SPIKEPro)

Turing AI Fellowship: PHOTONics for ultrafast Artificial Intelligence

Cite this