Physics-informed Neural-Network Software for Molecular Dynamics Applications

TL;DR

This paper introduces PND, a physics-informed neural network software tailored for molecular dynamics, enabling flexible simulation and optimization of physical laws through deep learning techniques.

Contribution

The paper presents a novel differential equation solver software, PND, that integrates physics-informed neural networks with a molecular dynamics engine for enhanced simulation flexibility.

Findings

Flexible implementation of physical laws as loss functions

Parallel MD engine accelerates PINN development

Enhanced optimization of conservation laws and boundary conditions

Abstract

We have developed a novel differential equation solver software called PND based on the physics-informed neural network for molecular dynamics simulators. Based on automatic differentiation technique provided by Pytorch, our software allows users to flexibly implement equation of atom motions, initial and boundary conditions, and conservation laws as loss function to train the network. PND comes with a parallel molecular dynamics (MD) engine in order for users to examine and optimize loss function design, and different conservation laws and boundary conditions, and hyperparameters, thereby accelerate the PINN-based development for molecular applications.