Reconfigurable platforms and the challenges for large-scale implementations of spiking neural networks

Jim Harkin; Fearghal Morgan; Steve Hall; Piotr Dudek; Thomas Dowrick; Liam McDaid

doi:10.1109/FPL.2008.4629989

Reconfigurable platforms and the challenges for large-scale implementations of spiking neural networks

Jim Harkin, Fearghal Morgan, Steve Hall, Piotr Dudek, Thomas Dowrick, Liam McDaid

Electrical & Electronic Engineering

Research output: Chapter in Book or Conference Publication/Proceeding › Conference Publication › peer-review

23 Citations (Scopus)

Abstract

FPGA devices have witnessed popularity in their use for the rapid prototyping of biological Spiking Neural Network (SNNs) applications, as they offer the key requirement of reconfigurability. However, FPGAs do not efficiently realise the biological neuron/synaptic models. Also their routing structures cannot accommodate the high levels of neuron inter-connectivity inherent in complex SNNs. This paper highlights and discusses the current challenges of implementing large scale SNNs on reconfigurable FPGAs. The paper presents a novel Field Programmable Neural Network (FPNN) architecture incorporating low power analogue synapse and a network on chip architecture for SNN routing and configuration. Initial results are presented.

Original language	English
Title of host publication	Proceedings - 2008 International Conference on Field Programmable Logic and Applications, FPL
Publisher	IEEE Computer Society
Pages	483-486
Number of pages	4
ISBN (Print)	9781424419616
DOIs	https://doi.org/10.1109/FPL.2008.4629989
Publication status	Published - 2008
Event	2008 International Conference on Field Programmable Logic and Applications, FPL - Heidelberg, Germany Duration: 8 Sep 2008 → 10 Sep 2008

Publication series

Name	Proceedings - 2008 International Conference on Field Programmable Logic and Applications, FPL

Conference

Conference	2008 International Conference on Field Programmable Logic and Applications, FPL
Country/Territory	Germany
City	Heidelberg
Period	8/09/08 → 10/09/08

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

10.1109/FPL.2008.4629989

Cite this

Harkin, J., Morgan, F., Hall, S., Dudek, P., Dowrick, T., & McDaid, L. (2008). Reconfigurable platforms and the challenges for large-scale implementations of spiking neural networks. In Proceedings - 2008 International Conference on Field Programmable Logic and Applications, FPL (pp. 483-486). Article 4629989 (Proceedings - 2008 International Conference on Field Programmable Logic and Applications, FPL). IEEE Computer Society. https://doi.org/10.1109/FPL.2008.4629989

Harkin, Jim ; Morgan, Fearghal ; Hall, Steve et al. / Reconfigurable platforms and the challenges for large-scale implementations of spiking neural networks. Proceedings - 2008 International Conference on Field Programmable Logic and Applications, FPL. IEEE Computer Society, 2008. pp. 483-486 (Proceedings - 2008 International Conference on Field Programmable Logic and Applications, FPL).

@inproceedings{385fee8959274fd3ac19fab1520d5c7e,

title = "Reconfigurable platforms and the challenges for large-scale implementations of spiking neural networks",

abstract = "FPGA devices have witnessed popularity in their use for the rapid prototyping of biological Spiking Neural Network (SNNs) applications, as they offer the key requirement of reconfigurability. However, FPGAs do not efficiently realise the biological neuron/synaptic models. Also their routing structures cannot accommodate the high levels of neuron inter-connectivity inherent in complex SNNs. This paper highlights and discusses the current challenges of implementing large scale SNNs on reconfigurable FPGAs. The paper presents a novel Field Programmable Neural Network (FPNN) architecture incorporating low power analogue synapse and a network on chip architecture for SNN routing and configuration. Initial results are presented.",

author = "Jim Harkin and Fearghal Morgan and Steve Hall and Piotr Dudek and Thomas Dowrick and Liam McDaid",

year = "2008",

doi = "10.1109/FPL.2008.4629989",

language = "English",

isbn = "9781424419616",

series = "Proceedings - 2008 International Conference on Field Programmable Logic and Applications, FPL",

publisher = "IEEE Computer Society",

pages = "483--486",

booktitle = "Proceedings - 2008 International Conference on Field Programmable Logic and Applications, FPL",

note = "2008 International Conference on Field Programmable Logic and Applications, FPL ; Conference date: 08-09-2008 Through 10-09-2008",

}

Harkin, J, Morgan, F, Hall, S, Dudek, P, Dowrick, T & McDaid, L 2008, Reconfigurable platforms and the challenges for large-scale implementations of spiking neural networks. in Proceedings - 2008 International Conference on Field Programmable Logic and Applications, FPL., 4629989, Proceedings - 2008 International Conference on Field Programmable Logic and Applications, FPL, IEEE Computer Society, pp. 483-486, 2008 International Conference on Field Programmable Logic and Applications, FPL, Heidelberg, Germany, 8/09/08. https://doi.org/10.1109/FPL.2008.4629989

Reconfigurable platforms and the challenges for large-scale implementations of spiking neural networks. / Harkin, Jim; Morgan, Fearghal; Hall, Steve et al.
Proceedings - 2008 International Conference on Field Programmable Logic and Applications, FPL. IEEE Computer Society, 2008. p. 483-486 4629989 (Proceedings - 2008 International Conference on Field Programmable Logic and Applications, FPL).

Research output: Chapter in Book or Conference Publication/Proceeding › Conference Publication › peer-review

TY - GEN

T1 - Reconfigurable platforms and the challenges for large-scale implementations of spiking neural networks

AU - Harkin, Jim

AU - Morgan, Fearghal

AU - Hall, Steve

AU - Dudek, Piotr

AU - Dowrick, Thomas

AU - McDaid, Liam

PY - 2008

Y1 - 2008

N2 - FPGA devices have witnessed popularity in their use for the rapid prototyping of biological Spiking Neural Network (SNNs) applications, as they offer the key requirement of reconfigurability. However, FPGAs do not efficiently realise the biological neuron/synaptic models. Also their routing structures cannot accommodate the high levels of neuron inter-connectivity inherent in complex SNNs. This paper highlights and discusses the current challenges of implementing large scale SNNs on reconfigurable FPGAs. The paper presents a novel Field Programmable Neural Network (FPNN) architecture incorporating low power analogue synapse and a network on chip architecture for SNN routing and configuration. Initial results are presented.

AB - FPGA devices have witnessed popularity in their use for the rapid prototyping of biological Spiking Neural Network (SNNs) applications, as they offer the key requirement of reconfigurability. However, FPGAs do not efficiently realise the biological neuron/synaptic models. Also their routing structures cannot accommodate the high levels of neuron inter-connectivity inherent in complex SNNs. This paper highlights and discusses the current challenges of implementing large scale SNNs on reconfigurable FPGAs. The paper presents a novel Field Programmable Neural Network (FPNN) architecture incorporating low power analogue synapse and a network on chip architecture for SNN routing and configuration. Initial results are presented.

UR - http://www.scopus.com/inward/record.url?scp=54949132181&partnerID=8YFLogxK

U2 - 10.1109/FPL.2008.4629989

DO - 10.1109/FPL.2008.4629989

M3 - Conference Publication

AN - SCOPUS:54949132181

SN - 9781424419616

T3 - Proceedings - 2008 International Conference on Field Programmable Logic and Applications, FPL

SP - 483

EP - 486

BT - Proceedings - 2008 International Conference on Field Programmable Logic and Applications, FPL

PB - IEEE Computer Society

T2 - 2008 International Conference on Field Programmable Logic and Applications, FPL

Y2 - 8 September 2008 through 10 September 2008

ER -

Harkin J, Morgan F, Hall S, Dudek P, Dowrick T, McDaid L. Reconfigurable platforms and the challenges for large-scale implementations of spiking neural networks. In Proceedings - 2008 International Conference on Field Programmable Logic and Applications, FPL. IEEE Computer Society. 2008. p. 483-486. 4629989. (Proceedings - 2008 International Conference on Field Programmable Logic and Applications, FPL). doi: 10.1109/FPL.2008.4629989

Reconfigurable platforms and the challenges for large-scale implementations of spiking neural networks

Abstract

Publication series

Conference

UN SDGs

Access to Document

Other files and links

Fingerprint

Cite this