Learning Low Rank Neural Representations of Hyperbolic Wave Dynamics from Data

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Cho, Woojin, Lee, Kookjin, Park, Noseong, Rim, Donsub, Welper, Gerrit
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866909885511237632
author Cho, Woojin
Lee, Kookjin
Park, Noseong
Rim, Donsub
Welper, Gerrit
author_facet Cho, Woojin
Lee, Kookjin
Park, Noseong
Rim, Donsub
Welper, Gerrit
contents We present a data-driven dimensionality reduction method that is well-suited for physics-based data representing hyperbolic wave propagation. The method utilizes a specialized neural network architecture called low rank neural representation (LRNR) inside a hypernetwork framework. The architecture is motivated by theoretical results that rigorously prove the existence of efficient representations for this wave class. We illustrate through archetypal examples that such an efficient low-dimensional representation of propagating waves can be learned directly from data through a combination of deep learning techniques. We observe that a low rank tensor representation arises naturally in the trained LRNRs, and that this reveals a new decomposition of wave propagation where each decomposed mode corresponds to interpretable physical features. Furthermore, we demonstrate that the LRNR architecture enables efficient inference via a compression scheme, which is a potentially important feature when deploying LRNRs in demanding performance regimes.
format Preprint
id arxiv_https___arxiv_org_abs_2510_25123
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Learning Low Rank Neural Representations of Hyperbolic Wave Dynamics from Data
Cho, Woojin
Lee, Kookjin
Park, Noseong
Rim, Donsub
Welper, Gerrit
Machine Learning
Artificial Intelligence
Numerical Analysis
68T07, 65D25, 65M22
We present a data-driven dimensionality reduction method that is well-suited for physics-based data representing hyperbolic wave propagation. The method utilizes a specialized neural network architecture called low rank neural representation (LRNR) inside a hypernetwork framework. The architecture is motivated by theoretical results that rigorously prove the existence of efficient representations for this wave class. We illustrate through archetypal examples that such an efficient low-dimensional representation of propagating waves can be learned directly from data through a combination of deep learning techniques. We observe that a low rank tensor representation arises naturally in the trained LRNRs, and that this reveals a new decomposition of wave propagation where each decomposed mode corresponds to interpretable physical features. Furthermore, we demonstrate that the LRNR architecture enables efficient inference via a compression scheme, which is a potentially important feature when deploying LRNRs in demanding performance regimes.
title Learning Low Rank Neural Representations of Hyperbolic Wave Dynamics from Data
topic Machine Learning
Artificial Intelligence
Numerical Analysis
68T07, 65D25, 65M22
url https://arxiv.org/abs/2510.25123