English

Rethinking Physics-Informed Regression Beyond Training Loops and Bespoke Architectures

Optimization and Control 2025-12-16 v1 Machine Learning

Abstract

We revisit the problem of physics-informed regression, and propose a method that directly computes the state at the prediction point, simultaneously with the derivative and curvature information of the existing samples. We frame each prediction as a constrained optimisation problem, leveraging multivariate Taylor series expansions and explicitly enforcing physical laws. Each individual query can be processed with low computational cost without any pre- or re-training, in contrast to global function approximator-based solutions such as neural networks. Our comparative benchmarks on a reaction-diffusion system show competitive predictive accuracy relative to a neural network-based solution, while completely eliminating the need for long training loops, and remaining robust to changes in the sampling layout.

Keywords

Cite

@article{arxiv.2512.13217,
  title  = {Rethinking Physics-Informed Regression Beyond Training Loops and Bespoke Architectures},
  author = {Lorenzo Sabug and Eric Kerrigan},
  journal= {arXiv preprint arXiv:2512.13217},
  year   = {2025}
}
R2 v1 2026-07-01T08:25:04.233Z