Do Machine-Learning Atomic Descriptors and Order Parameters Tell the Same Story? The Case of Liquid Water

Abstract

Machine-learning (ML) has become a key workhorse in molecular simulations. Building an ML model in this context, involves encoding the information of chemical environments using local atomic descriptors. In this work, we focus on the Smooth Overlap of Atomic Positions (SOAP) and their application in studying the properties of liquid water both in the bulk and at the hydrophobic air-water interface. By using a statistical test aimed at assessing the relative information content of different distance measures defined on the same data space, we investigate if these descriptors provide the same information as some of the common order parameters that are used to characterize local water structure such as hydrogen bonding, density or tetrahedrality to name a few. Our analysis suggests that the ML description and the standard order parameters of local water structure are not equivalent. In particular, a combination of these order parameters probing local water environments can predict SOAP similarity only approximately, and viceversa, the environments that are similar according to SOAP are not necessarily similar according to the standard order parameters. We also elucidate the role of some of the metaparameters entering in the SOAP definition in encoding chemical information.

References in corpus (8)

• Canonical sampling through velocity-rescaling
• ANI-1: An extensible neural network potential with DFT accuracy at force field computational cost
• PLUMED: a portable plugin for free-energy calculations with molecular dynamics
• Mixture-like behavior near a liquid-liquid phase transition in simulations of supercooled water
• Structure of the First and Second Neighbor Shells of Water: Quantitative Relation with Translational and Orientational Order
• Multiscale machine-learning interatomic potentials for ferromagnetic and liquid iron
• Ephemeral ice-like local environments in classical rigid models of liquid water
• The Conformational Space of a Flexible Amino Acid at Metallic Surfaces

Cited by in corpus (12)

• Beyond Local Structures In Critical Supercooled Water Through Unsupervised Learning
• ZundEig: The Structure of the Proton in Liquid Water From Unsupervised Learning
• "Layer-by-layer" Unsupervised Clustering of Statistically Relevant Fluctuations in Noisy Time-series Data of Complex Dynamical Systems
• Automatic feature selection and weighting in molecular systems using Differentiable Information Imbalance
• Data-driven discovery of statistically relevant information in quantum simulators
• Selecting Relevant Structural Features for Glassy Dynamics by Information Imbalance
• Aqueous Solution Chemistry In Silico and the Role of Data Driven Approaches
• Relevant, hidden, and frustrated information in high-dimensional analyses of complex dynamical systems with internal noise
• Data-driven assessment of optimal spatiotemporal resolutions for information extraction in noisy time series data
• A data driven approach to classify descriptors based on their efficiency in translating noisy trajectories into physically-relevant information
• Maximum Information Extraction Via Clustering and Minimization of Shannon Entropy
• Machine learning evaluation of structural descriptors for supercooled water