Impact of spiking neurons leakages and network recurrences on event-based spatio-temporal pattern recognition
Spiking neural networks coupled with neuromorphic hardware and event-based sensors are getting increased interest for low-latency and low-power inference at the edge. However, multiple spiking neuron models have been proposed in the literature with different levels of biological plausibility and dif...
Main Authors: | , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Frontiers Media S.A.
2023-11-01
|
Series: | Frontiers in Neuroscience |
Subjects: | |
Online Access: | https://www.frontiersin.org/articles/10.3389/fnins.2023.1244675/full |
_version_ | 1797465426568937472 |
---|---|
author | Mohamed Sadek Bouanane Dalila Cherifi Elisabetta Chicca Elisabetta Chicca Lyes Khacef Lyes Khacef |
author_facet | Mohamed Sadek Bouanane Dalila Cherifi Elisabetta Chicca Elisabetta Chicca Lyes Khacef Lyes Khacef |
author_sort | Mohamed Sadek Bouanane |
collection | DOAJ |
description | Spiking neural networks coupled with neuromorphic hardware and event-based sensors are getting increased interest for low-latency and low-power inference at the edge. However, multiple spiking neuron models have been proposed in the literature with different levels of biological plausibility and different computational features and complexities. Consequently, there is a need to define the right level of abstraction from biology in order to get the best performance in accurate, efficient and fast inference in neuromorphic hardware. In this context, we explore the impact of synaptic and membrane leakages in spiking neurons. We confront three neural models with different computational complexities using feedforward and recurrent topologies for event-based visual and auditory pattern recognition. Our results showed that, in terms of accuracy, leakages are important when there are both temporal information in the data and explicit recurrence in the network. Additionally, leakages do not necessarily increase the sparsity of spikes flowing in the network. We also investigated the impact of heterogeneity in the time constant of leakages. The results showed a slight improvement in accuracy when using data with a rich temporal structure, thereby validating similar findings obtained in previous studies. These results advance our understanding of the computational role of the neural leakages and network recurrences, and provide valuable insights for the design of compact and energy-efficient neuromorphic hardware for embedded systems. |
first_indexed | 2024-03-09T18:22:15Z |
format | Article |
id | doaj.art-2df37e9225f745259568443cf7ae123e |
institution | Directory Open Access Journal |
issn | 1662-453X |
language | English |
last_indexed | 2024-03-09T18:22:15Z |
publishDate | 2023-11-01 |
publisher | Frontiers Media S.A. |
record_format | Article |
series | Frontiers in Neuroscience |
spelling | doaj.art-2df37e9225f745259568443cf7ae123e2023-11-24T08:11:07ZengFrontiers Media S.A.Frontiers in Neuroscience1662-453X2023-11-011710.3389/fnins.2023.12446751244675Impact of spiking neurons leakages and network recurrences on event-based spatio-temporal pattern recognitionMohamed Sadek Bouanane0Dalila Cherifi1Elisabetta Chicca2Elisabetta Chicca3Lyes Khacef4Lyes Khacef5Institute of Electrical and Electronic Engineering, University of Boumerdes, Boumerdes, AlgeriaInstitute of Electrical and Electronic Engineering, University of Boumerdes, Boumerdes, AlgeriaBio-Inspired Circuits and Systems Lab, Zernike Institute for Advanced Materials, University of Groningen, Groningen, NetherlandsGroningen Cognitive Systems and Materials Center, University of Groningen, Groningen, NetherlandsBio-Inspired Circuits and Systems Lab, Zernike Institute for Advanced Materials, University of Groningen, Groningen, NetherlandsGroningen Cognitive Systems and Materials Center, University of Groningen, Groningen, NetherlandsSpiking neural networks coupled with neuromorphic hardware and event-based sensors are getting increased interest for low-latency and low-power inference at the edge. However, multiple spiking neuron models have been proposed in the literature with different levels of biological plausibility and different computational features and complexities. Consequently, there is a need to define the right level of abstraction from biology in order to get the best performance in accurate, efficient and fast inference in neuromorphic hardware. In this context, we explore the impact of synaptic and membrane leakages in spiking neurons. We confront three neural models with different computational complexities using feedforward and recurrent topologies for event-based visual and auditory pattern recognition. Our results showed that, in terms of accuracy, leakages are important when there are both temporal information in the data and explicit recurrence in the network. Additionally, leakages do not necessarily increase the sparsity of spikes flowing in the network. We also investigated the impact of heterogeneity in the time constant of leakages. The results showed a slight improvement in accuracy when using data with a rich temporal structure, thereby validating similar findings obtained in previous studies. These results advance our understanding of the computational role of the neural leakages and network recurrences, and provide valuable insights for the design of compact and energy-efficient neuromorphic hardware for embedded systems.https://www.frontiersin.org/articles/10.3389/fnins.2023.1244675/fullevent-based sensorsdigital neuromorphic architecturesspiking neural networksspatio-temporal patternsneurons leakagesneural heterogeneity |
spellingShingle | Mohamed Sadek Bouanane Dalila Cherifi Elisabetta Chicca Elisabetta Chicca Lyes Khacef Lyes Khacef Impact of spiking neurons leakages and network recurrences on event-based spatio-temporal pattern recognition Frontiers in Neuroscience event-based sensors digital neuromorphic architectures spiking neural networks spatio-temporal patterns neurons leakages neural heterogeneity |
title | Impact of spiking neurons leakages and network recurrences on event-based spatio-temporal pattern recognition |
title_full | Impact of spiking neurons leakages and network recurrences on event-based spatio-temporal pattern recognition |
title_fullStr | Impact of spiking neurons leakages and network recurrences on event-based spatio-temporal pattern recognition |
title_full_unstemmed | Impact of spiking neurons leakages and network recurrences on event-based spatio-temporal pattern recognition |
title_short | Impact of spiking neurons leakages and network recurrences on event-based spatio-temporal pattern recognition |
title_sort | impact of spiking neurons leakages and network recurrences on event based spatio temporal pattern recognition |
topic | event-based sensors digital neuromorphic architectures spiking neural networks spatio-temporal patterns neurons leakages neural heterogeneity |
url | https://www.frontiersin.org/articles/10.3389/fnins.2023.1244675/full |
work_keys_str_mv | AT mohamedsadekbouanane impactofspikingneuronsleakagesandnetworkrecurrencesoneventbasedspatiotemporalpatternrecognition AT dalilacherifi impactofspikingneuronsleakagesandnetworkrecurrencesoneventbasedspatiotemporalpatternrecognition AT elisabettachicca impactofspikingneuronsleakagesandnetworkrecurrencesoneventbasedspatiotemporalpatternrecognition AT elisabettachicca impactofspikingneuronsleakagesandnetworkrecurrencesoneventbasedspatiotemporalpatternrecognition AT lyeskhacef impactofspikingneuronsleakagesandnetworkrecurrencesoneventbasedspatiotemporalpatternrecognition AT lyeskhacef impactofspikingneuronsleakagesandnetworkrecurrencesoneventbasedspatiotemporalpatternrecognition |