Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Manor, Shalev, Kohandel, Mohammad
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:https://arxiv.org/abs/2505.16996
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866918030991163392
author Manor, Shalev
Kohandel, Mohammad
author_facet Manor, Shalev
Kohandel, Mohammad
contents Inverse problems involving differential equations often require identifying unknown parameters or functions from data. Existing approaches, such as Physics-Informed Neural Networks (PINNs), Universal Differential Equations (UDEs) and Universal Physics-Informed Neural Networks (UPINNs), are effective at isolating either parameters or functions but can face challenges when applied simultaneously due to solution non-uniqueness. In this work, we introduce a framework that addresses these limitations by establishing conditions under which unique solutions can be guaranteed. To illustrate, we apply it to examples from biological systems and ecological dynamics, demonstrating accurate and interpretable results. Our approach significantly enhances the potential of machine learning techniques in modeling complex systems in science and engineering.
format Preprint
id arxiv_https___arxiv_org_abs_2505_16996
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle A Unified Framework for Simultaneous Parameter and Function Discovery in Differential Equations
Manor, Shalev
Kohandel, Mohammad
Machine Learning
Inverse problems involving differential equations often require identifying unknown parameters or functions from data. Existing approaches, such as Physics-Informed Neural Networks (PINNs), Universal Differential Equations (UDEs) and Universal Physics-Informed Neural Networks (UPINNs), are effective at isolating either parameters or functions but can face challenges when applied simultaneously due to solution non-uniqueness. In this work, we introduce a framework that addresses these limitations by establishing conditions under which unique solutions can be guaranteed. To illustrate, we apply it to examples from biological systems and ecological dynamics, demonstrating accurate and interpretable results. Our approach significantly enhances the potential of machine learning techniques in modeling complex systems in science and engineering.
title A Unified Framework for Simultaneous Parameter and Function Discovery in Differential Equations
topic Machine Learning
url https://arxiv.org/abs/2505.16996