A theoretical analysis on the inversion of matrices via Neural Networks designed with Strassen algorithm

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Romera, Gonzalo, Bárcena-Petisco, Jon Asier
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866912878065352704
author Romera, Gonzalo
Bárcena-Petisco, Jon Asier
author_facet Romera, Gonzalo
Bárcena-Petisco, Jon Asier
contents We construct a Neural Network that approximates the matrix multiplication operator for any activation function such that there exists a Neural Network which can approximate the scalar multiplication function. In particular, we use the Strassen algorithm to reduce the number of weights and layers needed for such Neural Networks. This allows us to define another Neural Network for approximating the inverse matrix operator. Also, by relying on the Galerkin method, we apply those Neural Networks to solve parametric elliptic PDEs for a whole set of parameters. Finally, we discuss improvements with respect to the prior results.
format Preprint
id arxiv_https___arxiv_org_abs_2501_06539
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle A theoretical analysis on the inversion of matrices via Neural Networks designed with Strassen algorithm
Romera, Gonzalo
Bárcena-Petisco, Jon Asier
Numerical Analysis
Functional Analysis
35A35, 35J99, 41A25, 41A46, 65N30, 68T07
We construct a Neural Network that approximates the matrix multiplication operator for any activation function such that there exists a Neural Network which can approximate the scalar multiplication function. In particular, we use the Strassen algorithm to reduce the number of weights and layers needed for such Neural Networks. This allows us to define another Neural Network for approximating the inverse matrix operator. Also, by relying on the Galerkin method, we apply those Neural Networks to solve parametric elliptic PDEs for a whole set of parameters. Finally, we discuss improvements with respect to the prior results.
title A theoretical analysis on the inversion of matrices via Neural Networks designed with Strassen algorithm
topic Numerical Analysis
Functional Analysis
35A35, 35J99, 41A25, 41A46, 65N30, 68T07
url https://arxiv.org/abs/2501.06539