Deep Learning for Integro-Differential Elimination

Read full text Summarise with VIRAT
Authors: Lemaire, François, Louis Roussel
Cited by: 0
References: 23

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

We investigate the use of deep learning techniques for computing integral equations from systems of nonlinear differential equations.The class of integral equations we consider are very general and difficult to manipulate with purely symbolic computations. In order to rapidly explore and experiment with such complicated expressions, we have trained deep learning models to get valuable insight on the type of expressions that can be expected to be calculated using symbolic methods.The first contribution is a novel algorithm for converting an integral equation into a differential equation. The second is the discovery and resolution, via our trained models, of systems which are currently beyond the reach of classical computer algebra.

Fields of Study & Subjects

Polynomial and algebraic computation Mathematical and Computational Methods Model Reduction and Neural Networks

Publication details

Journal / source
HAL
Publication date
16 March 2026
Language
EN
Document type
Preprint