English

Label Propagation Training Schemes for Physics-Informed Neural Networks and Gaussian Processes

Machine Learning 2024-04-10 v1

Abstract

This paper proposes a semi-supervised methodology for training physics-informed machine learning methods. This includes self-training of physics-informed neural networks and physics-informed Gaussian processes in isolation, and the integration of the two via co-training. We demonstrate via extensive numerical experiments how these methods can ameliorate the issue of propagating information forward in time, which is a common failure mode of physics-informed machine learning.

Keywords

Cite

@article{arxiv.2404.05817,
  title  = {Label Propagation Training Schemes for Physics-Informed Neural Networks and Gaussian Processes},
  author = {Ming Zhong and Dehao Liu and Raymundo Arroyave and Ulisses Braga-Neto},
  journal= {arXiv preprint arXiv:2404.05817},
  year   = {2024}
}