Neural heterogeneity promotes robust learning
The authors show that heterogeneity in spiking neural networks improves accuracy and robustness of prediction for complex information processing tasks, results in optimal parameter distribution similar to experimental data and is metabolically efficient for learning tasks at varying timescales.
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Publishing Group
2021-10-01
|
| Series: | Nature Communications |
| Online Access: | https://doi.org/10.1038/s41467-021-26022-3 |
| id |
doaj-5f3aee1234744188899e09099851da31 |
|---|---|
| record_format |
Article |
| spelling |
doaj-5f3aee1234744188899e09099851da312021-10-10T11:45:03ZengNature Publishing GroupNature Communications2041-17232021-10-011211910.1038/s41467-021-26022-3Neural heterogeneity promotes robust learningNicolas Perez-Nieves0Vincent C. H. Leung1Pier Luigi Dragotti2Dan F. M. Goodman3Department of Electrical and Electronic Engineering, Imperial College LondonDepartment of Electrical and Electronic Engineering, Imperial College LondonDepartment of Electrical and Electronic Engineering, Imperial College LondonDepartment of Electrical and Electronic Engineering, Imperial College LondonThe authors show that heterogeneity in spiking neural networks improves accuracy and robustness of prediction for complex information processing tasks, results in optimal parameter distribution similar to experimental data and is metabolically efficient for learning tasks at varying timescales.https://doi.org/10.1038/s41467-021-26022-3 |
| collection |
DOAJ |
| language |
English |
| format |
Article |
| sources |
DOAJ |
| author |
Nicolas Perez-Nieves Vincent C. H. Leung Pier Luigi Dragotti Dan F. M. Goodman |
| spellingShingle |
Nicolas Perez-Nieves Vincent C. H. Leung Pier Luigi Dragotti Dan F. M. Goodman Neural heterogeneity promotes robust learning Nature Communications |
| author_facet |
Nicolas Perez-Nieves Vincent C. H. Leung Pier Luigi Dragotti Dan F. M. Goodman |
| author_sort |
Nicolas Perez-Nieves |
| title |
Neural heterogeneity promotes robust learning |
| title_short |
Neural heterogeneity promotes robust learning |
| title_full |
Neural heterogeneity promotes robust learning |
| title_fullStr |
Neural heterogeneity promotes robust learning |
| title_full_unstemmed |
Neural heterogeneity promotes robust learning |
| title_sort |
neural heterogeneity promotes robust learning |
| publisher |
Nature Publishing Group |
| series |
Nature Communications |
| issn |
2041-1723 |
| publishDate |
2021-10-01 |
| description |
The authors show that heterogeneity in spiking neural networks improves accuracy and robustness of prediction for complex information processing tasks, results in optimal parameter distribution similar to experimental data and is metabolically efficient for learning tasks at varying timescales. |
| url |
https://doi.org/10.1038/s41467-021-26022-3 |
| work_keys_str_mv |
AT nicolaspereznieves neuralheterogeneitypromotesrobustlearning AT vincentchleung neuralheterogeneitypromotesrobustlearning AT pierluigidragotti neuralheterogeneitypromotesrobustlearning AT danfmgoodman neuralheterogeneitypromotesrobustlearning |
| _version_ |
1716829590991667200 |