Learning on the Temporal Tangent Bundle for Physics-Informed Neural Networks

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Jamal, Adetola, Charbel, Mamlankou, Wilfrid, Houédanou Koffi, Guy, Dègla Aymard
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908960450150400
author Jamal, Adetola
Charbel, Mamlankou
Wilfrid, Houédanou Koffi
Guy, Dègla Aymard
author_facet Jamal, Adetola
Charbel, Mamlankou
Wilfrid, Houédanou Koffi
Guy, Dègla Aymard
contents This paper addresses the limitations of Physics-Informed Neural Networks for time-dependent problems by introducing a tangent bundle learning framework. Instead of directly approximating the solution, we parameterize its temporal derivative and reconstruct the state through a Volterra integral operator that enforces initial conditions exactly. This approach eliminates competing soft constraints and naturally amplifies high-frequency errors through differentiation, countering spectral bias. We prove theoretical equivalence between minimizing the differentiated residual and solving the original partial differential equation. Experiments on advection, Burgers, and Klein-Gordon equations show that the proposed method achieves 100 to 200 times lower errors than standard approaches using compact three-layer networks, with superior shock-capturing and long-time accuracy.
format Preprint
id arxiv_https___arxiv_org_abs_2604_11829
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Learning on the Temporal Tangent Bundle for Physics-Informed Neural Networks
Jamal, Adetola
Charbel, Mamlankou
Wilfrid, Houédanou Koffi
Guy, Dègla Aymard
Numerical Analysis
65M99, 35K05, 35Q53, 68T07
This paper addresses the limitations of Physics-Informed Neural Networks for time-dependent problems by introducing a tangent bundle learning framework. Instead of directly approximating the solution, we parameterize its temporal derivative and reconstruct the state through a Volterra integral operator that enforces initial conditions exactly. This approach eliminates competing soft constraints and naturally amplifies high-frequency errors through differentiation, countering spectral bias. We prove theoretical equivalence between minimizing the differentiated residual and solving the original partial differential equation. Experiments on advection, Burgers, and Klein-Gordon equations show that the proposed method achieves 100 to 200 times lower errors than standard approaches using compact three-layer networks, with superior shock-capturing and long-time accuracy.
title Learning on the Temporal Tangent Bundle for Physics-Informed Neural Networks
topic Numerical Analysis
65M99, 35K05, 35Q53, 68T07
url https://arxiv.org/abs/2604.11829