Blending Diverse Physical Priors with Neural Networks

TL;DR

This paper introduces PhysicsNAS, a neural architecture search method designed to enhance physics-based learning by effectively blending physical priors with neural networks, adaptable to various physical models and data qualities.

Contribution

The paper presents PhysicsNAS, the first neural architecture search approach tailored for physics-based learning, improving generalization across diverse physical priors and data conditions.

Findings

PhysicsNAS outperforms existing methods across multiple physical tasks.

It adapts effectively to variations in physical model accuracy.

Demonstrates robustness with different data qualities.

Abstract

Machine learning in context of physical systems merits a re-examination of the learning strategy. In addition to data, one can leverage a vast library of physical prior models (e.g. kinematics, fluid flow, etc) to perform more robust inference. The nascent sub-field of \emph{physics-based learning} (PBL) studies the blending of neural networks with physical priors. While previous PBL algorithms have been applied successfully to specific tasks, it is hard to generalize existing PBL methods to a wide range of physics-based problems. Such generalization would require an architecture that can adapt to variations in the correctness of the physics, or in the quality of training data. No such architecture exists. In this paper, we aim to generalize PBL, by making a first attempt to bring neural architecture search (NAS) to the realm of PBL. We introduce a new method known as physics-based neural architecture search (PhysicsNAS) that is a top-performer across a diverse range of quality in the physical model and the dataset.