Deep learning probability flows and entropy production rates in active matter

TL;DR

This paper introduces a deep learning approach using a novel transformer architecture to efficiently estimate entropy production and probability currents in high-dimensional active matter systems, enabling analysis of nonequilibrium states.

Contribution

The authors develop a spatially-local transformer network to estimate probability density scores, facilitating the computation of entropy production and probability currents in large active matter systems.

Findings

The method accurately estimates entropy production in high-dimensional active matter.

A single trained network generalizes across different system sizes and phase diagram regions.

The approach reveals the spatial structure of nonequilibrium behavior in motility-induced phase separation.

Abstract

Active matter systems, from self-propelled colloids to motile bacteria, are characterized by the conversion of free energy into useful work at the microscopic scale. They involve physics beyond the reach of equilibrium statistical mechanics, and a persistent challenge has been to understand the nature of their nonequilibrium states. The entropy production rate and the probability current provide quantitative ways to do so by measuring the breakdown of time-reversal symmetry. Yet, their efficient computation has remained elusive, as they depend on the system's unknown and high-dimensional probability density. Here, building upon recent advances in generative modeling, we develop a deep learning framework to estimate the score of this density. We show that the score, together with the microscopic equations of motion, gives access to the entropy production rate, the probability current, and their decomposition into local contributions from individual particles. To represent the score, we introduce a novel, spatially-local transformer network architecture that learns high-order interactions between particles while respecting their underlying permutation symmetry. We demonstrate the broad utility and scalability of the method by applying it to several high-dimensional systems of active particles undergoing motility-induced phase separation (MIPS). We show that a single network trained on a system of 4096 particles at one packing fraction can generalize to other regions of the phase diagram, including systems with as many as 32768 particles. We use this observation to quantify the spatial structure of the departure from equilibrium in MIPS as a function of the number of particles and the packing fraction.