Concept transfer of synaptic diversity from biological to artificial neural networks

Abstract Recent developments in artificial neural networks have drawn inspiration from biological neural networks, leveraging the concept of the artificial neuron to model the learning abilities of biological nerve cells. However, while neuroscience has provided new insights into the mechanisms of b...

Full description

Saved in:
Bibliographic Details
Main Authors: Martin Hofmann, Moritz Franz Peter Becker, Christian Tetzlaff, Patrick Mäder
Format: Article
Language:English
Published: Nature Portfolio 2025-06-01
Series:Nature Communications
Online Access:https://doi.org/10.1038/s41467-025-60078-9
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849725059034251264
author Martin Hofmann
Moritz Franz Peter Becker
Christian Tetzlaff
Patrick Mäder
author_facet Martin Hofmann
Moritz Franz Peter Becker
Christian Tetzlaff
Patrick Mäder
author_sort Martin Hofmann
collection DOAJ
description Abstract Recent developments in artificial neural networks have drawn inspiration from biological neural networks, leveraging the concept of the artificial neuron to model the learning abilities of biological nerve cells. However, while neuroscience has provided new insights into the mechanisms of biological neural networks, only a limited number of these concepts have been directly applied to artificial neural networks, with no guarantee of improved performance. Here, we address the discrepancy between the inhomogeneous and dynamic structures of biological neural networks and the largely homogeneous and fixed topologies of artificial neural networks. Specifically, we demonstrate successful integration of concepts of synaptic diversity, including spontaneous spine remodeling, synaptic plasticity diversity, and multi-synaptic connectivity, into artificial neural networks. Our findings reveal increased learning speed, prediction accuracy, and resilience to gradient inversion attacks. Our publicly available drop-in replacement code enables easy incorporation of these proposed concepts into existing networks.
format Article
id doaj-art-65c0f5dc48cf4d3e9db36401a9fdbbe7
institution DOAJ
issn 2041-1723
language English
publishDate 2025-06-01
publisher Nature Portfolio
record_format Article
series Nature Communications
spelling doaj-art-65c0f5dc48cf4d3e9db36401a9fdbbe72025-08-20T03:10:34ZengNature PortfolioNature Communications2041-17232025-06-0116111610.1038/s41467-025-60078-9Concept transfer of synaptic diversity from biological to artificial neural networksMartin Hofmann0Moritz Franz Peter Becker1Christian Tetzlaff2Patrick Mäder3Data-intensive Systems and Visualization Group (dAI.SY), Technische Universität IlmenauGroup of Computational Synaptic Physiology, Department for Neuro- and Sensory Physiology, University Medical Center GöttingenGroup of Computational Synaptic Physiology, Department for Neuro- and Sensory Physiology, University Medical Center GöttingenData-intensive Systems and Visualization Group (dAI.SY), Technische Universität IlmenauAbstract Recent developments in artificial neural networks have drawn inspiration from biological neural networks, leveraging the concept of the artificial neuron to model the learning abilities of biological nerve cells. However, while neuroscience has provided new insights into the mechanisms of biological neural networks, only a limited number of these concepts have been directly applied to artificial neural networks, with no guarantee of improved performance. Here, we address the discrepancy between the inhomogeneous and dynamic structures of biological neural networks and the largely homogeneous and fixed topologies of artificial neural networks. Specifically, we demonstrate successful integration of concepts of synaptic diversity, including spontaneous spine remodeling, synaptic plasticity diversity, and multi-synaptic connectivity, into artificial neural networks. Our findings reveal increased learning speed, prediction accuracy, and resilience to gradient inversion attacks. Our publicly available drop-in replacement code enables easy incorporation of these proposed concepts into existing networks.https://doi.org/10.1038/s41467-025-60078-9
spellingShingle Martin Hofmann
Moritz Franz Peter Becker
Christian Tetzlaff
Patrick Mäder
Concept transfer of synaptic diversity from biological to artificial neural networks
Nature Communications
title Concept transfer of synaptic diversity from biological to artificial neural networks
title_full Concept transfer of synaptic diversity from biological to artificial neural networks
title_fullStr Concept transfer of synaptic diversity from biological to artificial neural networks
title_full_unstemmed Concept transfer of synaptic diversity from biological to artificial neural networks
title_short Concept transfer of synaptic diversity from biological to artificial neural networks
title_sort concept transfer of synaptic diversity from biological to artificial neural networks
url https://doi.org/10.1038/s41467-025-60078-9
work_keys_str_mv AT martinhofmann concepttransferofsynapticdiversityfrombiologicaltoartificialneuralnetworks
AT moritzfranzpeterbecker concepttransferofsynapticdiversityfrombiologicaltoartificialneuralnetworks
AT christiantetzlaff concepttransferofsynapticdiversityfrombiologicaltoartificialneuralnetworks
AT patrickmader concepttransferofsynapticdiversityfrombiologicaltoartificialneuralnetworks