Dendrites endow artificial neural networks with accurate, robust and parameter-efficient learning
Abstract Artificial neural networks (ANNs) are at the core of most Deep Learning (DL) algorithms that successfully tackle complex problems like image recognition, autonomous driving, and natural language processing. However, unlike biological brains who tackle similar problems in a very efficient ma...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2025-01-01
|
| Series: | Nature Communications |
| Online Access: | https://doi.org/10.1038/s41467-025-56297-9 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832585585023254528 |
|---|---|
| author | Spyridon Chavlis Panayiota Poirazi |
| author_facet | Spyridon Chavlis Panayiota Poirazi |
| author_sort | Spyridon Chavlis |
| collection | DOAJ |
| description | Abstract Artificial neural networks (ANNs) are at the core of most Deep Learning (DL) algorithms that successfully tackle complex problems like image recognition, autonomous driving, and natural language processing. However, unlike biological brains who tackle similar problems in a very efficient manner, DL algorithms require a large number of trainable parameters, making them energy-intensive and prone to overfitting. Here, we show that a new ANN architecture that incorporates the structured connectivity and restricted sampling properties of biological dendrites counteracts these limitations. We find that dendritic ANNs are more robust to overfitting and match or outperform traditional ANNs on several image classification tasks while using significantly fewer trainable parameters. These advantages are likely the result of a different learning strategy, whereby most of the nodes in dendritic ANNs respond to multiple classes, unlike classical ANNs that strive for class-specificity. Our findings suggest that the incorporation of dendritic properties can make learning in ANNs more precise, resilient, and parameter-efficient and shed new light on how biological features can impact the learning strategies of ANNs. |
| format | Article |
| id | doaj-art-aba7a47ddb70485bb4f28343941dadf6 |
| institution | Kabale University |
| issn | 2041-1723 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Nature Communications |
| spelling | doaj-art-aba7a47ddb70485bb4f28343941dadf62025-01-26T12:42:19ZengNature PortfolioNature Communications2041-17232025-01-0116111710.1038/s41467-025-56297-9Dendrites endow artificial neural networks with accurate, robust and parameter-efficient learningSpyridon Chavlis0Panayiota Poirazi1Institute of Molecular Biology and Biotechnology, Foundation for Research and Technology-HellasInstitute of Molecular Biology and Biotechnology, Foundation for Research and Technology-HellasAbstract Artificial neural networks (ANNs) are at the core of most Deep Learning (DL) algorithms that successfully tackle complex problems like image recognition, autonomous driving, and natural language processing. However, unlike biological brains who tackle similar problems in a very efficient manner, DL algorithms require a large number of trainable parameters, making them energy-intensive and prone to overfitting. Here, we show that a new ANN architecture that incorporates the structured connectivity and restricted sampling properties of biological dendrites counteracts these limitations. We find that dendritic ANNs are more robust to overfitting and match or outperform traditional ANNs on several image classification tasks while using significantly fewer trainable parameters. These advantages are likely the result of a different learning strategy, whereby most of the nodes in dendritic ANNs respond to multiple classes, unlike classical ANNs that strive for class-specificity. Our findings suggest that the incorporation of dendritic properties can make learning in ANNs more precise, resilient, and parameter-efficient and shed new light on how biological features can impact the learning strategies of ANNs.https://doi.org/10.1038/s41467-025-56297-9 |
| spellingShingle | Spyridon Chavlis Panayiota Poirazi Dendrites endow artificial neural networks with accurate, robust and parameter-efficient learning Nature Communications |
| title | Dendrites endow artificial neural networks with accurate, robust and parameter-efficient learning |
| title_full | Dendrites endow artificial neural networks with accurate, robust and parameter-efficient learning |
| title_fullStr | Dendrites endow artificial neural networks with accurate, robust and parameter-efficient learning |
| title_full_unstemmed | Dendrites endow artificial neural networks with accurate, robust and parameter-efficient learning |
| title_short | Dendrites endow artificial neural networks with accurate, robust and parameter-efficient learning |
| title_sort | dendrites endow artificial neural networks with accurate robust and parameter efficient learning |
| url | https://doi.org/10.1038/s41467-025-56297-9 |
| work_keys_str_mv | AT spyridonchavlis dendritesendowartificialneuralnetworkswithaccuraterobustandparameterefficientlearning AT panayiotapoirazi dendritesendowartificialneuralnetworkswithaccuraterobustandparameterefficientlearning |