To main content

PIGPVAE: physics-informed gaussian process variational autoencoders

Abstract

Recent advances in generative AI offer promising solutions for synthetic data generation but often rely on large datasets for effective training. To address this limitation, we propose a novel generative model that learns from limited data by incorporating physical constraints into a Variational Autoencoder (VAE) framework. Specifically, we extend VAE with a physics-based generator to capture underlying dynamics, while unmodeled dynamics are learned via a latent Gaussian Process VAE (GPVAE) component. We further introduce a regularization term that balances the physical model and data-driven discrepancy, promoting both interpretability and fidelity to real-world observations. We evaluate the proposed method on both real and simulated data, demonstrating that the Physics-Informed GPVAE (PIGPVAE) outperforms state-of-the-art methods in terms of diversity and accuracy of the generated samples, even under small-data conditions. Additionally, we demonstrate that PIGPVAE can produce realistic samples beyond the observed distribution, highlighting its robustness and usefulness under distribution shifts.
Read the publication

Category

Academic article

Language

English

Author(s)

Affiliation

  • SINTEF Community / Architecture, Materials and Structures
  • SINTEF Digital / Software Engineering, Safety and Security
  • Norwegian University of Science and Technology

Date

28.07.2025

Year

2025

Published in

Applied intelligence (Boston)

ISSN

0924-669X

Volume

55

Issue

12

View this publication at Norwegian Research Information Repository