Information and Configurational Entropy in Glassy Systems

TL;DR

This paper proposes a novel method to quantify glassiness in molecular systems by using lossless compression of concatenated snapshots, providing a direct measure of vibrational and configurational entropies and insights into glass transition phenomena.

Contribution

It introduces a new entropy measurement technique based on data compression, linking information theory with glass physics to analyze structural motifs and glass transition behavior.

Findings

Compressed file size correlates with vibrational and configurational entropy.

Method can detect divergence of glass length scale at transition.

Applicable to systems out of equilibrium.

Abstract

It is often stated that if one is presented with a snapshot of the positions of the molecules of a glass and one of a liquid, one is unable to tell the difference. Here we argue instead that given several such snapshots taken over a time-interval, even without specifying the times, there is a definite procedure to assess precisely the level of glassiness: it suffices to concatenate the snapshots side-by-side, and to subject the joint picture to a lossless compression protocol. We argue that the size of the compressed file yields a direct and unambiguous measure of the `vibrational' and `configurational' entropies, and may be used to study the associated glass length scale in or out of equilibrium through the size and frequency of the repeated motifs essential to the compression, a quantity that would diverge at a putative glass transition.