From 47d3f0fc6e2327a1cc912f4c548fbe49f9a8f78a Mon Sep 17 00:00:00 2001 From: Thomas Keck Date: Mon, 6 Apr 2020 15:05:25 +0000 Subject: [PATCH] Update abstract in README.md of glassy_dynamics. PiperOrigin-RevId: 305029846 --- README.md | 1 + glassy_dynamics/README.md | 29 +++++++++++++++-------------- 2 files changed, 16 insertions(+), 14 deletions(-) diff --git a/README.md b/README.md index c68ae03..1cc904b 100644 --- a/README.md +++ b/README.md @@ -24,6 +24,7 @@ https://deepmind.com/research/publications/ ## Projects +* [Unveiling the predictive power of static structure in glassy systems](glassy_dynamics), Nature Physics 2020 * [Multi-Object Representation Learning with Iterative Variational Inference (IODINE)](iodine) * [AlphaFold CASP13](alphafold_casp13), Nature 2020 * [Unrestricted Adversarial Challenge](unrestricted_advx) diff --git a/glassy_dynamics/README.md b/glassy_dynamics/README.md index 482718a..bff9cf4 100644 --- a/glassy_dynamics/README.md +++ b/glassy_dynamics/README.md @@ -1,7 +1,8 @@ # Unveiling the predictive power of static structure in glassy systems This repository contains an open source implementation of the graph neural -network model described in our paper. +network model described in our +[paper](http://dx.doi.org/10.1038/s41567-020-0842-8). The model can be trained using the training binary included in this repository, and the dataset published with our paper. @@ -9,18 +10,18 @@ and the dataset published with our paper. ## Abstract Despite decades of theoretical studies, the nature of the glass transition -remains elusive and debated, while the existence of structural predictors of the +remains elusive and debated, while the existence of structural predictors of its dynamics is a major open question. Recent approaches propose inferring predictors from a variety of human-defined features using machine learning. -We learn the long time evolution of a glassy system solely from the initial -particle positions and without any hand-crafted features, using a powerful -model: graph neural networks. We show that this method strongly outperforms -state-of-the-art methods, generalizing over a wide range of temperatures, -pressures, and densities. In shear experiments, it predicts the location of -rearranging particles. The structural predictors learned by our network unveil a -correlation length which increases with larger timescales to reach the size of -our system. Beyond glasses, our method could apply to many other physical -systems that map to a graph of local interactions. +Here we determine the long time evolution of a glassy system solely from the +initial particle positions and without any hand-crafted features, using graph +neural networks as a powerful model. We show that this method outperforms +current state-of-the-art methods, generalizing over a wide range of +temperatures, pressures, and densities. In shear experiments, it predicts the +locations of rearranging particles. The structural predictors learned by our +network exhibit a correlation length which increases with larger timescales to +reach the size of our system. Beyond glasses, our method could apply to many +other physical systems that map to a graph of local interaction. ## Dataset @@ -70,9 +71,9 @@ crossed a periodic boundary during the simulation. If this repository is helpful for your research please cite the following publication: -Unveiling the predictive power of static structure in glassysystems -V. Bapst, T. Keck, A. Grabska-Barwinska, C. Donner, E. D. Cubuk, -S. S. Schoenholz, A.Obika, A. W. R. Nelson, T. Back, D. Hassabis and P. Kohli +[Unveiling the predictive power of static structure in glassy systems](http://dx.doi.org/10.1038/s41567-020-0842-8) +V. Bapst, T. Keck, A. Grabska-Barwińska, C. Donner, E. D. Cubuk, +S. S. Schoenholz, A. Obika, A. W. R. Nelson, T. Back, D. Hassabis and P. Kohli ## Disclaimer