Investigation into the pragmatism of phenomenological spiking neurons for hardware implementation on FPGAs

S Johnston, G Prasad, LP Maguire, TM McGinnity, A Belatreche

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Spiking neurons (SNs) are biologicallyplausible neuron models that offer new informationprocessing paradigms for neuroengineers. It is expectedthat artificial representation of these neurons willenhance the link between biological and artificialsystems. The complexity of spiking neuron models withlow level abstraction makes them unsuitable for largescale implementations, limiting network scalability. Thishas led to the development of simpler, phenomenologicalspike models, such as the Leaky Integrate and Fire model.However, no clear guidelines exist to help select whichphenomenological model to implement. The aim of thispaper is to reduce this ambiguity, through a systematiccomparative performance evaluation. An evolutionarystrategy for the supervised training of networks to twoformal models is used to solve computational benchmarkproblems in software. The models are then designed,simulated and implemented onto a Field ProgrammableGate Array (FPGA) through a novel hardware designflow. It is envisaged that this information will helpneuroengineers in future hardware implementationdecisions.
LanguageEnglish
Title of host publicationUnknown Host Publication
Pages90-95
Number of pages6
Publication statusPublished - Sep 2004
EventIEEE SMC UK-RI chapter Conference - Derry
Duration: 1 Sep 2004 → …

Conference

ConferenceIEEE SMC UK-RI chapter Conference
Period1/09/04 → …

Fingerprint

Neurons
Field programmable gate arrays (FPGA)
Hardware
Scalability

Cite this

@inproceedings{1ce33749e5804da18936e547af70fe75,
title = "Investigation into the pragmatism of phenomenological spiking neurons for hardware implementation on FPGAs",
abstract = "Spiking neurons (SNs) are biologicallyplausible neuron models that offer new informationprocessing paradigms for neuroengineers. It is expectedthat artificial representation of these neurons willenhance the link between biological and artificialsystems. The complexity of spiking neuron models withlow level abstraction makes them unsuitable for largescale implementations, limiting network scalability. Thishas led to the development of simpler, phenomenologicalspike models, such as the Leaky Integrate and Fire model.However, no clear guidelines exist to help select whichphenomenological model to implement. The aim of thispaper is to reduce this ambiguity, through a systematiccomparative performance evaluation. An evolutionarystrategy for the supervised training of networks to twoformal models is used to solve computational benchmarkproblems in software. The models are then designed,simulated and implemented onto a Field ProgrammableGate Array (FPGA) through a novel hardware designflow. It is envisaged that this information will helpneuroengineers in future hardware implementationdecisions.",
author = "S Johnston and G Prasad and LP Maguire and TM McGinnity and A Belatreche",
year = "2004",
month = "9",
language = "English",
pages = "90--95",
booktitle = "Unknown Host Publication",

}

Johnston, S, Prasad, G, Maguire, LP, McGinnity, TM & Belatreche, A 2004, Investigation into the pragmatism of phenomenological spiking neurons for hardware implementation on FPGAs. in Unknown Host Publication. pp. 90-95, IEEE SMC UK-RI chapter Conference, 1/09/04.

Investigation into the pragmatism of phenomenological spiking neurons for hardware implementation on FPGAs. / Johnston, S; Prasad, G; Maguire, LP; McGinnity, TM; Belatreche, A.

Unknown Host Publication. 2004. p. 90-95.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - Investigation into the pragmatism of phenomenological spiking neurons for hardware implementation on FPGAs

AU - Johnston, S

AU - Prasad, G

AU - Maguire, LP

AU - McGinnity, TM

AU - Belatreche, A

PY - 2004/9

Y1 - 2004/9

N2 - Spiking neurons (SNs) are biologicallyplausible neuron models that offer new informationprocessing paradigms for neuroengineers. It is expectedthat artificial representation of these neurons willenhance the link between biological and artificialsystems. The complexity of spiking neuron models withlow level abstraction makes them unsuitable for largescale implementations, limiting network scalability. Thishas led to the development of simpler, phenomenologicalspike models, such as the Leaky Integrate and Fire model.However, no clear guidelines exist to help select whichphenomenological model to implement. The aim of thispaper is to reduce this ambiguity, through a systematiccomparative performance evaluation. An evolutionarystrategy for the supervised training of networks to twoformal models is used to solve computational benchmarkproblems in software. The models are then designed,simulated and implemented onto a Field ProgrammableGate Array (FPGA) through a novel hardware designflow. It is envisaged that this information will helpneuroengineers in future hardware implementationdecisions.

AB - Spiking neurons (SNs) are biologicallyplausible neuron models that offer new informationprocessing paradigms for neuroengineers. It is expectedthat artificial representation of these neurons willenhance the link between biological and artificialsystems. The complexity of spiking neuron models withlow level abstraction makes them unsuitable for largescale implementations, limiting network scalability. Thishas led to the development of simpler, phenomenologicalspike models, such as the Leaky Integrate and Fire model.However, no clear guidelines exist to help select whichphenomenological model to implement. The aim of thispaper is to reduce this ambiguity, through a systematiccomparative performance evaluation. An evolutionarystrategy for the supervised training of networks to twoformal models is used to solve computational benchmarkproblems in software. The models are then designed,simulated and implemented onto a Field ProgrammableGate Array (FPGA) through a novel hardware designflow. It is envisaged that this information will helpneuroengineers in future hardware implementationdecisions.

M3 - Conference contribution

SP - 90

EP - 95

BT - Unknown Host Publication

ER -