Unsupervised learning for local structure detection in colloidal systems

Abstract

We introduce a simple, fast, and easy to implement unsupervised learning algorithm for detecting different local environments on a single-particle level in colloidal systems. In this algorithm, we use a vector of standard bond-orientational order parameters to describe the local environment of each particle. We then use a neural-network-based autoencoder combined with Gaussian mixture models in order to autonomously group together similar environments. We test the performance of the method on snapshots of a wide variety of colloidal systems obtained via computer simulations, ranging from simple isotropically interacting systems to binary mixtures, and even anisotropic hard cubes. Additionally, we look at a variety of common self-assembled situations such as fluid-crystal and crystal-crystal coexistences, grain boundaries, and nucleation. In all cases, we are able to identify the relevant local environments to a similar precision as "standard," manually tuned, and system-specific, order parameters. In addition to classifying such environments, we also use the trained autoencoder in order to determine the most relevant bond orientational order parameters in the systems analyzed.

Autonomously revealing hidden local structures in supercooled liquids

“…The UML method we explore here is based on an algorithm we recently developed 10 for detecting crystalline structures. As shown in Fig.…”

“…The UML approach used here follows the method outlined in ref. 10 . A detailed description is provided in the Supplementary Methods .…”

“…Here we show that a simple, efficient UML algorithm that we recently developed 10 for detecting crystalline structure can be harnessed to detect structural heterogeneities in glasses. Our algorithm—which requires only a single snapshot as input—uses bond-order parameters (BOPs) to encode the local environments of the particles.…”

Characterising soft matter using machine learning

“…In a new departure, Boattini and coworkers used a neuralnetwork based autoencoder to create a compact representation of the bond order around each site. 59 An autoencoder begins life as two neural networks, the first (the encoder) performs a dimensional reduction and the second neural network (the decoder) takes this compact representation and expands it again, Fig. 3(b).…”

“…Bond order parameters are combined with an autoencoder to provide a compact description of the particle sites. 59 The compact description is then the basis of an unsupervised division of the sites into two classes. For both binary hard spheres and Wahnstrom glasses, the probability of being in one of the two classes of site is very highly correlated with the propensity; for the Kob-Anderson glass the correlation is not quite as strong.…”