MetaPhysiCa: OOD Robustness in Physics-Informed Machine Learning
Abstract
A fundamental challenge in physics-informed machine learning (PIML) is the design of robust PIML methods for out-of-distribution (OOD) forecasting tasks. These OOD tasks require learning-to-learn from observations of the same (ODE) dynamical system with different unknown ODE parameters, and demand accurate forecasts even under out-of-support initial conditions and out-of-support ODE parameters. We propose a solution for such tasks, defined as a meta-learning procedure for causal structure discovery. In 3 different OOD tasks, we show that the proposed approach outperforms existing PIML and deep learning methods.
Cite
Text
Mouli et al. "MetaPhysiCa: OOD Robustness in Physics-Informed Machine Learning." ICLR 2023 Workshops: Physics4ML, 2023.Markdown
[Mouli et al. "MetaPhysiCa: OOD Robustness in Physics-Informed Machine Learning." ICLR 2023 Workshops: Physics4ML, 2023.](https://mlanthology.org/iclrw/2023/mouli2023iclrw-metaphysica/)BibTeX
@inproceedings{mouli2023iclrw-metaphysica,
title = {{MetaPhysiCa: OOD Robustness in Physics-Informed Machine Learning}},
author = {Mouli, S Chandra and Alam, Muhammad and Ribeiro, Bruno},
booktitle = {ICLR 2023 Workshops: Physics4ML},
year = {2023},
url = {https://mlanthology.org/iclrw/2023/mouli2023iclrw-metaphysica/}
}