Machine learning glass caging order parameters with an artificial nested neural network

TL;DR

This paper introduces a nested neural network approach to learn particle caging features directly from dynamics, effectively identifying phase transitions in glassy systems without relying on structural predictors.

Contribution

It presents a novel machine learning method that directly learns dynamical caging order parameters for multiple glass transitions, including melting, Gardner, and liquid to glass transitions.

Findings

Successfully identifies phase transitions in simulated glass models.

Distinguishes between true phase transitions and crossovers.

Provides a general approach for learning dynamical features in glassy systems.

Abstract

Around a glass transition, the dynamics of a supercooled liquid dramatically slow down, exhibited by caging of particles, while the structural changes remain subtle. In alternative to recent machine learning studies searching for structural predictors of glassy dynamics, here we propose to learn directly particle caging features defined purely according to dynamics. We focus on three transitions in a simulated hard sphere glass model, the melting of ultra-stable glasses, the Gardner transition and the liquid to ordinary glass transition. Implementing the machine learning algorithm based on a two-level nested neural network, we attain not only proper caging order parameters for all three transitions, but also a phase classification for input samples. A finite-size scaling analysis of the phase classification results identifies the order of melting (first) and Gardner (second) transitions. A false positive is avoided, as the liquid to glass transition is indicated as a crossover, rather than a phase transition with a well-defined transition point. This study paves the way to a generic approach for learning dynamical features in glassy systems, with a minimum requirement of system-specific knowledge.