Machine Learning for Phase Behavior in Active Matter Systems

TL;DR

This paper shows that deep learning models, including graph neural networks, can accurately predict phase separation in active matter systems at the particle level, aligning well with simulation results.

Contribution

It introduces a novel approach using deep learning to predict phase behavior in active matter, enabling particle-level phase classification and phase diagram determination.

Findings

Deep learning accurately predicts phase separation in active Brownian particles.

Graph neural networks effectively classify particle phases based on features.

Predictions strongly agree with traditional simulation-based phase diagrams.

Abstract

We demonstrate that deep learning techniques can be used to predict motility induced phase separation (MIPS) in suspensions of active Brownian particles (ABPs) by creating a notion of phase at the particle level. Using a fully connected network in conjunction with a graph neural network we use individual particle features to predict to which phase a particle belongs. From this, we are able to compute the fraction of dilute particles to determine if the system is in the homogeneous dilute, dense, or coexistence region. Our predictions are compared against the MIPS binodal computed from simulation. The strong agreement between the two suggests that machine learning provides an effective way to determine the phase behavior of ABPs and could prove useful for determining more complex phase diagrams.