Characterizing possible failure modes in physics-informed neural networks (characterizing-pinns-failure-modes) v0.1.0

Abstract

This software provides an implementation of the physics-informed neural network setup for different differential equations. It provides a way to solve differential equations using machine learning, via a physical constraint term in the loss function. Physics-informed neural networks are a popular approach, but here, the failures of such an approach are characterized and better solutions and training procedures are provided.