Inductive biases of neural network modularity in spatial navigation

dc.citation.articleNumbereadk1256en_US
dc.citation.issueNumber29en_US
dc.citation.journalTitleScience Advancesen_US
dc.citation.volumeNumber10en_US
dc.contributor.authorZhang, Ruiyien_US
dc.contributor.authorPitkow, Xaqen_US
dc.contributor.authorAngelaki, Dora E.en_US
dc.date.accessioned2024-08-09T16:25:26Zen_US
dc.date.available2024-08-09T16:25:26Zen_US
dc.date.issued2024en_US
dc.description.abstractThe brain may have evolved a modular architecture for daily tasks, with circuits featuring functionally specialized modules that match the task structure. We hypothesize that this architecture enables better learning and generalization than architectures with less specialized modules. To test this, we trained reinforcement learning agents with various neural architectures on a naturalistic navigation task. We found that the modular agent, with an architecture that segregates computations of state representation, value, and action into specialized modules, achieved better learning and generalization. Its learned state representation combines prediction and observation, weighted by their relative uncertainty, akin to recursive Bayesian estimation. This agent’s behavior also resembles macaques’ behavior more closely. Our results shed light on the possible rationale for the brain’s modularity and suggest that artificial systems can use this insight from neuroscience to improve learning and generalization in natural tasks.en_US
dc.identifier.citationZhang, R., Pitkow, X., & Angelaki, D. E. (2024). Inductive biases of neural network modularity in spatial navigation. Science Advances, 10(29), eadk1256. https://doi.org/10.1126/sciadv.adk1256en_US
dc.identifier.digitalsciadv-adk1256en_US
dc.identifier.doihttps://doi.org/10.1126/sciadv.adk1256en_US
dc.identifier.urihttps://hdl.handle.net/1911/117647en_US
dc.language.isoengen_US
dc.publisherAAASen_US
dc.rightsExcept where otherwise noted, this work is licensed under a Creative Commons Attribution (CC BY) license.  Permission to reuse, publish, or reproduce the work beyond the terms of the license or beyond the bounds of fair use or other exemptions to copyright law must be obtained from the copyright holder.en_US
dc.rights.urihttps://creativecommons.org/licenses/by/4.0/en_US
dc.titleInductive biases of neural network modularity in spatial navigationen_US
dc.typeJournal articleen_US
dc.type.dcmiTexten_US
dc.type.publicationpublisher versionen_US
Files
Original bundle
Now showing 1 - 1 of 1
Loading...
Thumbnail Image
Name:
sciadv-adk1256.pdf
Size:
2.16 MB
Format:
Adobe Portable Document Format