CONFIDE: Contextual Finite Differences Modelling of PDEs

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Linial, Ori, Avner, Orly, Di Castro, Dotan
Natura: Preprint
Pubblicazione: 2023
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866910475792416768
author Linial, Ori
Avner, Orly
Di Castro, Dotan
author_facet Linial, Ori
Avner, Orly
Di Castro, Dotan
contents We introduce a method for inferring an explicit PDE from a data sample generated by previously unseen dynamics, based on a learned context. The training phase integrates knowledge of the form of the equation with a differential scheme, while the inference phase yields a PDE that fits the data sample and enables both signal prediction and data explanation. We include results of extensive experimentation, comparing our method to SOTA approaches, together with ablation studies that examine different flavors of our solution.
format Preprint
id arxiv_https___arxiv_org_abs_2303_15827
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle CONFIDE: Contextual Finite Differences Modelling of PDEs
Linial, Ori
Avner, Orly
Di Castro, Dotan
Machine Learning
Numerical Analysis
We introduce a method for inferring an explicit PDE from a data sample generated by previously unseen dynamics, based on a learned context. The training phase integrates knowledge of the form of the equation with a differential scheme, while the inference phase yields a PDE that fits the data sample and enables both signal prediction and data explanation. We include results of extensive experimentation, comparing our method to SOTA approaches, together with ablation studies that examine different flavors of our solution.
title CONFIDE: Contextual Finite Differences Modelling of PDEs
topic Machine Learning
Numerical Analysis
url https://arxiv.org/abs/2303.15827