Emergent Riemannian geometry over learning discrete computations on continuous manifolds

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Brandon, Julian, Chadwick, Angus, Pellegrino, Arthur
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917113481920512
author Brandon, Julian
Chadwick, Angus
Pellegrino, Arthur
author_facet Brandon, Julian
Chadwick, Angus
Pellegrino, Arthur
contents Many tasks require mapping continuous input data (e.g. images) to discrete task outputs (e.g. class labels). Yet, how neural networks learn to perform such discrete computations on continuous data manifolds remains poorly understood. Here, we show that signatures of such computations emerge in the representational geometry of neural networks as they learn. By analysing the Riemannian pullback metric across layers of a neural network, we find that network computation can be decomposed into two functions: discretising continuous input features and performing logical operations on these discretised variables. Furthermore, we demonstrate how different learning regimes (rich vs. lazy) have contrasting metric and curvature structures, affecting the ability of the networks to generalise to unseen inputs. Overall, our work provides a geometric framework for understanding how neural networks learn to perform discrete computations on continuous manifolds.
format Preprint
id arxiv_https___arxiv_org_abs_2512_00196
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Emergent Riemannian geometry over learning discrete computations on continuous manifolds
Brandon, Julian
Chadwick, Angus
Pellegrino, Arthur
Machine Learning
Neural and Evolutionary Computing
Differential Geometry
Neurons and Cognition
Many tasks require mapping continuous input data (e.g. images) to discrete task outputs (e.g. class labels). Yet, how neural networks learn to perform such discrete computations on continuous data manifolds remains poorly understood. Here, we show that signatures of such computations emerge in the representational geometry of neural networks as they learn. By analysing the Riemannian pullback metric across layers of a neural network, we find that network computation can be decomposed into two functions: discretising continuous input features and performing logical operations on these discretised variables. Furthermore, we demonstrate how different learning regimes (rich vs. lazy) have contrasting metric and curvature structures, affecting the ability of the networks to generalise to unseen inputs. Overall, our work provides a geometric framework for understanding how neural networks learn to perform discrete computations on continuous manifolds.
title Emergent Riemannian geometry over learning discrete computations on continuous manifolds
topic Machine Learning
Neural and Evolutionary Computing
Differential Geometry
Neurons and Cognition
url https://arxiv.org/abs/2512.00196