Physics-Enhanced Machine Learning: a position paper for dynamical systems investigations

TL;DR

This paper discusses the development and challenges of Physics-Enhanced Machine Learning strategies tailored for dynamical systems, emphasizing their role in improving data efficiency, interpretability, and decision-making in engineering contexts.

Contribution

It provides a comprehensive overview of PEML approaches, defining the field, categorizing strategies, and analyzing their advantages and challenges for complex dynamical systems.

Findings

PEML strategies improve data efficiency and interpretability.

Three main PEML approaches: physics-guided, physics-encoded, physics-informed.

Challenges include integrating physics with ML and ensuring reliable decision-making.

Abstract

This position paper takes a broad look at Physics-Enhanced Machine Learning (PEML) -- also known as Scientific Machine Learning -- with particular focus to those PEML strategies developed to tackle dynamical systems' challenges. The need to go beyond Machine Learning (ML) strategies is driven by: (i) limited volume of informative data, (ii) avoiding accurate-but-wrong predictions; (iii) dealing with uncertainties; (iv) providing Explainable and Interpretable inferences. A general definition of PEML is provided by considering four physics and domain knowledge biases, and three broad groups of PEML approaches are discussed: physics-guided, physics-encoded and physics-informed. The advantages and challenges in developing PEML strategies for guiding high-consequence decision making in engineering applications involving complex dynamical systems, are presented.