Cargando…

Biologically plausible local synaptic learning rules robustly implement deep supervised learning

In deep neural networks, representational learning in the middle layer is essential for achieving efficient learning. However, the currently prevailing backpropagation learning rules (BP) are not necessarily biologically plausible and cannot be implemented in the brain in their current form. Therefo...

Descripción completa

Detalles Bibliográficos
Autores principales: Konishi, Masataka, Igarashi, Kei M., Miura, Keiji
Formato: Online Artículo Texto
Lenguaje:English
Publicado: Frontiers Media S.A. 2023
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10598703/
https://www.ncbi.nlm.nih.gov/pubmed/37886676
http://dx.doi.org/10.3389/fnins.2023.1160899
_version_ 1785125612855230464
author Konishi, Masataka
Igarashi, Kei M.
Miura, Keiji
author_facet Konishi, Masataka
Igarashi, Kei M.
Miura, Keiji
author_sort Konishi, Masataka
collection PubMed
description In deep neural networks, representational learning in the middle layer is essential for achieving efficient learning. However, the currently prevailing backpropagation learning rules (BP) are not necessarily biologically plausible and cannot be implemented in the brain in their current form. Therefore, to elucidate the learning rules used by the brain, it is critical to establish biologically plausible learning rules for practical memory tasks. For example, learning rules that result in a learning performance worse than that of animals observed in experimental studies may not be computations used in real brains and should be ruled out. Using numerical simulations, we developed biologically plausible learning rules to solve a task that replicates a laboratory experiment where mice learned to predict the correct reward amount. Although the extreme learning machine (ELM) and weight perturbation (WP) learning rules performed worse than the mice, the feedback alignment (FA) rule achieved a performance equal to that of BP. To obtain a more biologically plausible model, we developed a variant of FA, FA_Ex-100%, which implements direct dopamine inputs that provide error signals locally in the layer of focus, as found in the mouse entorhinal cortex. The performance of FA_Ex-100% was comparable to that of conventional BP. Finally, we tested whether FA_Ex-100% was robust against rule perturbations and biologically inevitable noise. FA_Ex-100% worked even when subjected to perturbations, presumably because it could calibrate the correct prediction error (e.g., dopaminergic signals) in the next step as a teaching signal if the perturbation created a deviation. These results suggest that simplified and biologically plausible learning rules, such as FA_Ex-100%, can robustly facilitate deep supervised learning when the error signal, possibly conveyed by dopaminergic neurons, is accurate.
format Online
Article
Text
id pubmed-10598703
institution National Center for Biotechnology Information
language English
publishDate 2023
publisher Frontiers Media S.A.
record_format MEDLINE/PubMed
spelling pubmed-105987032023-10-26 Biologically plausible local synaptic learning rules robustly implement deep supervised learning Konishi, Masataka Igarashi, Kei M. Miura, Keiji Front Neurosci Neuroscience In deep neural networks, representational learning in the middle layer is essential for achieving efficient learning. However, the currently prevailing backpropagation learning rules (BP) are not necessarily biologically plausible and cannot be implemented in the brain in their current form. Therefore, to elucidate the learning rules used by the brain, it is critical to establish biologically plausible learning rules for practical memory tasks. For example, learning rules that result in a learning performance worse than that of animals observed in experimental studies may not be computations used in real brains and should be ruled out. Using numerical simulations, we developed biologically plausible learning rules to solve a task that replicates a laboratory experiment where mice learned to predict the correct reward amount. Although the extreme learning machine (ELM) and weight perturbation (WP) learning rules performed worse than the mice, the feedback alignment (FA) rule achieved a performance equal to that of BP. To obtain a more biologically plausible model, we developed a variant of FA, FA_Ex-100%, which implements direct dopamine inputs that provide error signals locally in the layer of focus, as found in the mouse entorhinal cortex. The performance of FA_Ex-100% was comparable to that of conventional BP. Finally, we tested whether FA_Ex-100% was robust against rule perturbations and biologically inevitable noise. FA_Ex-100% worked even when subjected to perturbations, presumably because it could calibrate the correct prediction error (e.g., dopaminergic signals) in the next step as a teaching signal if the perturbation created a deviation. These results suggest that simplified and biologically plausible learning rules, such as FA_Ex-100%, can robustly facilitate deep supervised learning when the error signal, possibly conveyed by dopaminergic neurons, is accurate. Frontiers Media S.A. 2023-10-11 /pmc/articles/PMC10598703/ /pubmed/37886676 http://dx.doi.org/10.3389/fnins.2023.1160899 Text en Copyright © 2023 Konishi, Igarashi and Miura. https://creativecommons.org/licenses/by/4.0/This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
spellingShingle Neuroscience
Konishi, Masataka
Igarashi, Kei M.
Miura, Keiji
Biologically plausible local synaptic learning rules robustly implement deep supervised learning
title Biologically plausible local synaptic learning rules robustly implement deep supervised learning
title_full Biologically plausible local synaptic learning rules robustly implement deep supervised learning
title_fullStr Biologically plausible local synaptic learning rules robustly implement deep supervised learning
title_full_unstemmed Biologically plausible local synaptic learning rules robustly implement deep supervised learning
title_short Biologically plausible local synaptic learning rules robustly implement deep supervised learning
title_sort biologically plausible local synaptic learning rules robustly implement deep supervised learning
topic Neuroscience
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10598703/
https://www.ncbi.nlm.nih.gov/pubmed/37886676
http://dx.doi.org/10.3389/fnins.2023.1160899
work_keys_str_mv AT konishimasataka biologicallyplausiblelocalsynapticlearningrulesrobustlyimplementdeepsupervisedlearning
AT igarashikeim biologicallyplausiblelocalsynapticlearningrulesrobustlyimplementdeepsupervisedlearning
AT miurakeiji biologicallyplausiblelocalsynapticlearningrulesrobustlyimplementdeepsupervisedlearning