Advancing spatio-temporal processing through adaptation in spiking neural networks

Abstract Implementations of spiking neural networks on neuromorphic hardware promise orders of magnitude less power consumption than their non-spiking counterparts. The standard neuron model for spike-based computation on such systems has long been the leaky integrate-and-fire neuron. A computationa...

Full description

Saved in:
Bibliographic Details
Main Authors: Maximilian Baronig, Romain Ferrand, Silvester Sabathiel, Robert Legenstein
Format: Article
Language:English
Published: Nature Portfolio 2025-07-01
Series:Nature Communications
Online Access:https://doi.org/10.1038/s41467-025-60878-z
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849238404644995072
author Maximilian Baronig
Romain Ferrand
Silvester Sabathiel
Robert Legenstein
author_facet Maximilian Baronig
Romain Ferrand
Silvester Sabathiel
Robert Legenstein
author_sort Maximilian Baronig
collection DOAJ
description Abstract Implementations of spiking neural networks on neuromorphic hardware promise orders of magnitude less power consumption than their non-spiking counterparts. The standard neuron model for spike-based computation on such systems has long been the leaky integrate-and-fire neuron. A computationally light augmentation of this neuron model with an adaptation mechanism has recently been shown to exhibit superior performance on spatio-temporal processing tasks. The root of the superiority of these so-called adaptive leaky integrate-and-fire neurons however is not well understood. In this article, we thoroughly analyze the dynamical, computational, and learning properties of adaptive leaky integrate-and-fire neurons and networks thereof. Our investigation reveals significant challenges related to stability and parameterization when employing the conventional Euler-Forward discretization for this class of models. We report a rigorous theoretical and empirical demonstration that these challenges can be effectively addressed by adopting an alternative discretization approach – the Symplectic Euler method, allowing to improve over state-of-the-art performances on common event-based benchmark datasets. Our further analysis of the computational properties of these networks shows that they are particularly well suited to exploit the spatio-temporal structure of input sequences without any normalization techniques.
format Article
id doaj-art-bc9cbc81f41345288e4ed3b6009511bf
institution Kabale University
issn 2041-1723
language English
publishDate 2025-07-01
publisher Nature Portfolio
record_format Article
series Nature Communications
spelling doaj-art-bc9cbc81f41345288e4ed3b6009511bf2025-08-20T04:01:36ZengNature PortfolioNature Communications2041-17232025-07-0116112610.1038/s41467-025-60878-zAdvancing spatio-temporal processing through adaptation in spiking neural networksMaximilian Baronig0Romain Ferrand1Silvester Sabathiel2Robert Legenstein3Institute of Machine Learning and Neural Computation, Graz University of TechnologyInstitute of Machine Learning and Neural Computation, Graz University of TechnologySilicon Austria Labs GmbHInstitute of Machine Learning and Neural Computation, Graz University of TechnologyAbstract Implementations of spiking neural networks on neuromorphic hardware promise orders of magnitude less power consumption than their non-spiking counterparts. The standard neuron model for spike-based computation on such systems has long been the leaky integrate-and-fire neuron. A computationally light augmentation of this neuron model with an adaptation mechanism has recently been shown to exhibit superior performance on spatio-temporal processing tasks. The root of the superiority of these so-called adaptive leaky integrate-and-fire neurons however is not well understood. In this article, we thoroughly analyze the dynamical, computational, and learning properties of adaptive leaky integrate-and-fire neurons and networks thereof. Our investigation reveals significant challenges related to stability and parameterization when employing the conventional Euler-Forward discretization for this class of models. We report a rigorous theoretical and empirical demonstration that these challenges can be effectively addressed by adopting an alternative discretization approach – the Symplectic Euler method, allowing to improve over state-of-the-art performances on common event-based benchmark datasets. Our further analysis of the computational properties of these networks shows that they are particularly well suited to exploit the spatio-temporal structure of input sequences without any normalization techniques.https://doi.org/10.1038/s41467-025-60878-z
spellingShingle Maximilian Baronig
Romain Ferrand
Silvester Sabathiel
Robert Legenstein
Advancing spatio-temporal processing through adaptation in spiking neural networks
Nature Communications
title Advancing spatio-temporal processing through adaptation in spiking neural networks
title_full Advancing spatio-temporal processing through adaptation in spiking neural networks
title_fullStr Advancing spatio-temporal processing through adaptation in spiking neural networks
title_full_unstemmed Advancing spatio-temporal processing through adaptation in spiking neural networks
title_short Advancing spatio-temporal processing through adaptation in spiking neural networks
title_sort advancing spatio temporal processing through adaptation in spiking neural networks
url https://doi.org/10.1038/s41467-025-60878-z
work_keys_str_mv AT maximilianbaronig advancingspatiotemporalprocessingthroughadaptationinspikingneuralnetworks
AT romainferrand advancingspatiotemporalprocessingthroughadaptationinspikingneuralnetworks
AT silvestersabathiel advancingspatiotemporalprocessingthroughadaptationinspikingneuralnetworks
AT robertlegenstein advancingspatiotemporalprocessingthroughadaptationinspikingneuralnetworks