Exact Fixed-Point Constraints in Neural-ODEs with Provable Universality

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Pacifico, Feliciano Giuseppe, Fanelli, Duccio, Buffoni, Lorenzo, Chicchi, Lorenzo, Febbe, Diego, Marino, Raffaele
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909032953937920
author Pacifico, Feliciano Giuseppe
Fanelli, Duccio
Buffoni, Lorenzo
Chicchi, Lorenzo
Febbe, Diego
Marino, Raffaele
author_facet Pacifico, Feliciano Giuseppe
Fanelli, Duccio
Buffoni, Lorenzo
Chicchi, Lorenzo
Febbe, Diego
Marino, Raffaele
contents We introduce a technique that enables Neural-ODEs to approximate arbitrary velocity fields with a priori planted fixed-points. Specifically, a recipe is given to explicitly accommodate for a finite collection of points in the reference multi-dimensional space of the Neural-ODE where the velocity field is exactly equal to zero. In this way, the gradient-based training is rigorously constrained inside the prescribed hypothesis class while leaving the expressive power of the Neural-ODE unaltered. We rigorously prove the universality of the Neural-ODE under any local constraints in the velocity field and give a computationally convenient way of imposing the fixed points. Our method is then tested on two paradigmatic physical models.
format Preprint
id arxiv_https___arxiv_org_abs_2605_10613
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Exact Fixed-Point Constraints in Neural-ODEs with Provable Universality
Pacifico, Feliciano Giuseppe
Fanelli, Duccio
Buffoni, Lorenzo
Chicchi, Lorenzo
Febbe, Diego
Marino, Raffaele
Disordered Systems and Neural Networks
Machine Learning
We introduce a technique that enables Neural-ODEs to approximate arbitrary velocity fields with a priori planted fixed-points. Specifically, a recipe is given to explicitly accommodate for a finite collection of points in the reference multi-dimensional space of the Neural-ODE where the velocity field is exactly equal to zero. In this way, the gradient-based training is rigorously constrained inside the prescribed hypothesis class while leaving the expressive power of the Neural-ODE unaltered. We rigorously prove the universality of the Neural-ODE under any local constraints in the velocity field and give a computationally convenient way of imposing the fixed points. Our method is then tested on two paradigmatic physical models.
title Exact Fixed-Point Constraints in Neural-ODEs with Provable Universality
topic Disordered Systems and Neural Networks
Machine Learning
url https://arxiv.org/abs/2605.10613