Identifying structural flow defects in disordered solids using machine-learning methods.

By Ekin Dogus Cubuk, Sam Schoenholz, Jennifer M. Rieser, B. Malone, Joerg Rottler, D. Durian, Efthimios Kaxiras, A. J. Liu. 2014

336 citations recorded by the source corpus.

We use machine-learning methods on local structure to identify flow defects-or particles susceptible to rearrangement-in jammed and glassy systems. We apply this method successfully to two very different systems: a two-dimensional experimental realization of a granular pillar under compression and a Lennard-Jones glass in both two and three dimensions above and below its glass transition temperature. We also identify characteristics of flow defects that differentiate them from the rest of the sample. Our results show it is possible to discern subtle structural features responsible for heterogeneous dynamics observed across a broad range of disordered materials.

Subfields: Materials Chemistry, Mechanical Engineering

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Materials Chemistry

In Mechanical Engineering

Citations in the corpus

JAX, M.D.: End-to-End Differentiable, Hardware Accelerated, Molecular Dynamics in Pure Python (2019, methodology): “Coupled with the growing interest in deep learning, machine learning (ML) has become a popular tool for analyzing data that is produced by MD (Cubuk et al., 2015; Schoenholz et al., 2016; Cubuk et al., 2016; 2017b; Schoenholz, 2018; Rajak et al., 2019b;a; Sharp et al., 2018; Sussman et al., 2017;…”
End-to-End Differentiable , Hardware Accelerated , Molecular Dynamics in Pure Python (2019, methodology): “Coupled with the growing interest in deep learning, machine learning (ML) has become a popular tool for analyzing data that is produced by MD (Cubuk et al., 2015; Schoenholz et al., 2016; Cubuk et al., 2016; Schoenholz et al., 2017; Cubuk et al., 2017b; Schoenholz, 2018; Rajak et al., 2019b,a;…”
Predicting plasticity in disordered solids from structural indicators (2020, positive): “(b) Machine learning-based methods [35, 37–39] correlate a high-dimensional description of a local particle environment (constructed from purely structural information) with the likelihood of undergoing a local shear transformation.”
Roadmap on multiscale materials modeling (2020, background): “Computer algorithms will naturally be part of this endeavor, in terms of machine-learning algorithms for pattern recognition [99].”
Intriguing Properties of Adversarial Examples (2017, background): “…and engineering ﬁelds, from speech recognition (Mikolov et al., 2011; Hinton et al., 2012) to computer vision (Krizhevsky et al., 2012; Szegedy et al., 2015), and has provided novel insights in well-established ﬁelds such as condensed matter physics (Cubuk et al., 2015; Schoenholz et al., 2016a).”
Machine learning and the physical sciences (2019, background): “Disordered and glassy solids where identifications of the order parameter is particularly challenging were studied in (Cubuk et al., 2015; Schoenholz et al., 2017).”
Deformation and flow of amorphous solids: Insights from elastoplastic models (2017, background): “5c. Leaving behind traditional approaches, a couple of recent papers showed that it is possible to train an algorithm to recognize the atomic-scale patterns characteristic of a glassy state and spot its ‘soft’ regions (Cubuk et al., 2015, 2016; Schoenholz et al., 2016).”