Saved in:
Bibliographic Details
Main Authors: Leitao, Álvaro, Ráfales, Jonatan
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2512.11530
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912760038686720
author Leitao, Álvaro
Ráfales, Jonatan
author_facet Leitao, Álvaro
Ráfales, Jonatan
contents In this work, we introduce a machine/deep learning methodology to solve parametric integrals. Besides classical machine learning approaches, we consider a differential learning framework that incorporates derivative information during training, emphasizing its advantageous properties. Our study covers three representative problem classes: statistical functionals (including moments and cumulative distribution functions), approximation of functions via Chebyshev expansions, and integrals arising directly from differential equations. These examples range from smooth closed-form benchmarks to challenging numerical integrals. Across all cases, the differential machine learning-based approach consistently outperforms standard architectures, achieving lower mean squared error, enhanced scalability, and improved sample efficiency.
format Preprint
id arxiv_https___arxiv_org_abs_2512_11530
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Parametric Numerical Integration with (Differential) Machine Learning
Leitao, Álvaro
Ráfales, Jonatan
Machine Learning
Numerical Analysis
68T07, 65D30, 65C05
G.1.4; G.3
In this work, we introduce a machine/deep learning methodology to solve parametric integrals. Besides classical machine learning approaches, we consider a differential learning framework that incorporates derivative information during training, emphasizing its advantageous properties. Our study covers three representative problem classes: statistical functionals (including moments and cumulative distribution functions), approximation of functions via Chebyshev expansions, and integrals arising directly from differential equations. These examples range from smooth closed-form benchmarks to challenging numerical integrals. Across all cases, the differential machine learning-based approach consistently outperforms standard architectures, achieving lower mean squared error, enhanced scalability, and improved sample efficiency.
title Parametric Numerical Integration with (Differential) Machine Learning
topic Machine Learning
Numerical Analysis
68T07, 65D30, 65C05
G.1.4; G.3
url https://arxiv.org/abs/2512.11530