Learning reveals invisible structure in low-rank RNNs
Abstract
Learning in neural systems arises from synaptic changes that reshape the representations underlying behavior. While low-rank recurrent neural networks (RNNs) have emerged as a powerful framework for linking connectivity to function, a theoretical understanding of their learning process remains elusive. Here, we extend the low-rank framework from activity to learning by deriving gradient-descent dynamics directly in a reduced overlap space. We formulate a closed-form, low-dimensional system of ODEs that governs learning in this space, exact for linear RNNs and asymptotically exact for nonlinear RNNs in the large-N Gaussian limit. Central to our analysis is a distinction between two classes of overlaps: loss-visible overlaps, which fully determine network activity, output, and loss, and loss-invisible overlaps, which do not affect function but are required to describe learning. We illustrate the consequences of this decomposition through two phenomena. First, we show that learning can serve as a perturbation that exposes differences in connectivity between functionally equivalent networks. Second, we show that loss-invisible overlaps can act as memory variables that encode training history, and characterize the conditions under which this occurs. Finally, we present several testable predictions for biological learning experiments derived from our theory.
Related
- CUBE: Multimodal Representation Learning Reveals Biological Structure Across Histomorphology, Spatial Protein Phenotypes, and Transcriptome-Associated Signals
- Prediction emerges in RNNs trained for perception
- A foundation for systematic analysis of transformers and RNNs for tractography
- Oscillatory Phase Coding Emerges in Adaptive RNNS for Flexible Working Memory
- Oscillatory Phase Coding Emerges in Adaptive RNNs for Flexible Working Memory
BibTeX
@misc{ger2026learning,
title = {{Learning reveals invisible structure in low-rank RNNs}},
author = {Yoav Ger and Omri Barak},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2605.04115},
eprint = {2605.04115},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2605.04115},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Learning reveals invisible structure in low-rank RNNs
AU - Yoav Ger
AU - Omri Barak
PY - 2026
DA - 2026/10/16
PB - arXiv
DO - 10.48550/arxiv.2605.04115
UR - https://arxiv.org/abs/2605.04115
AB - Learning in neural systems arises from synaptic changes that reshape the representations underlying behavior. While low-rank recurrent neural networks (RNNs) have emerged as a powerful framework for linking connectivity to function, a theoretical understanding of their learning process remains elusive. Here, we extend the low-rank framework from activity to learning by deriving gradient-descent dynamics directly in a reduced overlap space. We formulate a closed-form, low-dimensional system of ODEs that governs learning in this space, exact for linear RNNs and asymptotically exact for nonlinear RNNs in the large-N Gaussian limit. Central to our analysis is a distinction between two classes of overlaps: loss-visible overlaps, which fully determine network activity, output, and loss, and loss-invisible overlaps, which do not affect function but are required to describe learning. We illustrate the consequences of this decomposition through two phenomena. First, we show that learning can serve as a perturbation that exposes differences in connectivity between functionally equivalent networks. Second, we show that loss-invisible overlaps can act as memory variables that encode training history, and characterize the conditions under which this occurs. Finally, we present several testable predictions for biological learning experiments derived from our theory.
ER -