English

Learning Low Rank Neural Representations of Hyperbolic Wave Dynamics from Data

Machine Learning 2025-11-05 v2 Artificial Intelligence Numerical Analysis Numerical Analysis

Abstract

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.

Keywords

Cite

@article{arxiv.2510.25123,
  title  = {Learning Low Rank Neural Representations of Hyperbolic Wave Dynamics from Data},
  author = {Woojin Cho and Kookjin Lee and Noseong Park and Donsub Rim and Gerrit Welper},
  journal= {arXiv preprint arXiv:2510.25123},
  year   = {2025}
}

Comments

41 pages, 18 figures

R2 v1 2026-07-01T07:10:58.070Z