Sufficient conditions for offline reactivation in recurrent neural networks

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Krishna, Nanda H., Bredenberg, Colin, Levenstein, Daniel, Richards, Blake A., Lajoie, Guillaume
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915298914861056
author Krishna, Nanda H.
Bredenberg, Colin
Levenstein, Daniel
Richards, Blake A.
Lajoie, Guillaume
author_facet Krishna, Nanda H.
Bredenberg, Colin
Levenstein, Daniel
Richards, Blake A.
Lajoie, Guillaume
contents During periods of quiescence, such as sleep, neural activity in many brain circuits resembles that observed during periods of task engagement. However, the precise conditions under which task-optimized networks can autonomously reactivate the same network states responsible for online behavior is poorly understood. In this study, we develop a mathematical framework that outlines sufficient conditions for the emergence of neural reactivation in circuits that encode features of smoothly varying stimuli. We demonstrate mathematically that noisy recurrent networks optimized to track environmental state variables using change-based sensory information naturally develop denoising dynamics, which, in the absence of input, cause the network to revisit state configurations observed during periods of online activity. We validate our findings using numerical experiments on two canonical neuroscience tasks: spatial position estimation based on self-motion cues, and head direction estimation based on angular velocity cues. Overall, our work provides theoretical support for modeling offline reactivation as an emergent consequence of task optimization in noisy neural circuits.
format Preprint
id arxiv_https___arxiv_org_abs_2505_17003
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Sufficient conditions for offline reactivation in recurrent neural networks
Krishna, Nanda H.
Bredenberg, Colin
Levenstein, Daniel
Richards, Blake A.
Lajoie, Guillaume
Neurons and Cognition
Machine Learning
Neural and Evolutionary Computing
During periods of quiescence, such as sleep, neural activity in many brain circuits resembles that observed during periods of task engagement. However, the precise conditions under which task-optimized networks can autonomously reactivate the same network states responsible for online behavior is poorly understood. In this study, we develop a mathematical framework that outlines sufficient conditions for the emergence of neural reactivation in circuits that encode features of smoothly varying stimuli. We demonstrate mathematically that noisy recurrent networks optimized to track environmental state variables using change-based sensory information naturally develop denoising dynamics, which, in the absence of input, cause the network to revisit state configurations observed during periods of online activity. We validate our findings using numerical experiments on two canonical neuroscience tasks: spatial position estimation based on self-motion cues, and head direction estimation based on angular velocity cues. Overall, our work provides theoretical support for modeling offline reactivation as an emergent consequence of task optimization in noisy neural circuits.
title Sufficient conditions for offline reactivation in recurrent neural networks
topic Neurons and Cognition
Machine Learning
Neural and Evolutionary Computing
url https://arxiv.org/abs/2505.17003