Learning from data has led to paradigm shifts in a multitude of disciplines, including web, text and image search, speech recognition, as well as bioinformatics. Can machine learning enable similar breakthroughs in understanding quantum many-body systems? Here we develop an efficient deep learning approach that enables spatially and chemically resolved insights into quantum-mechanical observables of molecular systems. We unify concepts from many-body Hamiltonians with purpose-designed deep tensor neural networks, which leads to size-extensive and uniformly accurate (1 kcal mol−1) predictions in compositional and configurational chemical space for molecules of intermediate size. As an example of chemical relevance, the model reveals a classification of aromatic rings with respect to their stability. Further applications of our model for predicting atomic energies and local chemical potentials in molecules, reliable isomer energies, and molecules with peculiar electronic structure demonstrate the potential of machine learning for revealing insights into complex quantum-chemical systems.

/pdf/quantum-chemical-insights-from-deep-tensor-neural-networks-266u885pa4.pdf

Quantum-Chemical Insights from Deep Tensor Neural Networks

Atomic-scale modeling and understanding of materials have made remarkable progress, but they are still fundamentally limited by the large computational cost of explicit electronic-structure methods such as density-functional theory. This Progress Report shows how machine learning (ML) is currently enabling a new degree of realism in materials modeling: by "learning" electronic-structure data, ML-based interatomic potentials give access to atomistic simulations that reach similar accuracy levels but are orders of magnitude faster. A brief introduction to the new tools is given, and then, applications to some select problems in materials science are highlighted: phase-change materials for memory devices; nanoparticle catalysts; and carbon-based electrodes for chemical sensing, supercapacitors, and batteries. It is hoped that the present work will inspire the development and wider use of ML-based interatomic potentials in diverse areas of materials research.

/pdf/machine-learning-interatomic-potentials-as-emerging-tools-55nyropwu8.pdf

Machine Learning Interatomic Potentials as Emerging Tools for Materials Science.

https://pure.mpg.de/pubman/item/item_3314332_3/component/file_3324953/1-s2.0-S0370157320300363-main.pdf

A perspective on conventional high-temperature superconductors at high pressure: Methods and materials

We introduce a scheme for molecular simulations, the deep potential molecular dynamics (DPMD) method, based on a many-body potential and interatomic forces generated by a carefully crafted deep neural network trained with ab initio data. The neural network model preserves all the natural symmetries in the problem. It is first-principles based in the sense that there are no ad hoc components aside from the network model. We show that the proposed scheme provides an efficient and accurate protocol in a variety of systems, including bulk materials and molecules. In all these cases, DPMD gives results that are essentially indistinguishable from the original data, at a cost that scales linearly with system size.

Deep Potential Molecular Dynamics: a Scalable Model with the Accuracy of Quantum Mechanics

We provide an introduction to Gaussian process regression (GPR) machine-learning methods in computational materials science and chemistry. The focus of the present review is on the regression of atomistic properties: in particular, on the construction of interatomic potentials, or force fields, in the Gaussian Approximation Potential (GAP) framework; beyond this, we also discuss the fitting of arbitrary scalar, vectorial, and tensorial quantities. Methodological aspects of reference data generation, representation, and regression, as well as the question of how a data-driven model may be validated, are reviewed and critically discussed. A survey of applications to a variety of research questions in chemistry and materials science illustrates the rapid growth in the field. A vision is outlined for the development of the methodology in the years to come.

Gaussian Process Regression for Materials and Molecules.

Ab initio structure prediction methods have been nowadays widely used as powerful tools for structure searches and materials discovery. However, they are generally restricted to small systems owing to the heavy computational cost of the underlying density functional theory (DFT) calculations in structure optimizations. In this work, by combining a state-of-art machine learning (ML) potential with our in-house developed CALYPSO structure prediction method, we developed two acceleration schemes for the structure prediction of large systems, in which a ML potential is pre-constructed to fully replace DFT calculations or trained in an on-the-fly manner from scratch during the structure searches. The developed schemes have been applied to medium- and large-sized boron clusters, both of which are challenging cases for either the construction of ML potentials or extensive structure searches. Experimental structures of B36 and B40 clusters can be readily reproduced, and the putative global minimum structure for the B84 cluster is proposed, where the computational cost is substantially reduced by ∼1-2 orders of magnitude if compared with full DFT-based structure searches. Our results demonstrate a viable route for structure prediction in large systems via the combination of state-of-art structure prediction methods and ML techniques.

Accelerating CALYPSO structure prediction by data-driven learning of a potential energy surface

Ab initio structure prediction methods have been nowadays widely used as powerful tools for structure searches and material discovery However, they are generally restricted to small systems owing to the heavy computational cost of underlying density functional theory (DFT) calculations In this work, by combining state-of-art machine learning (ML) potential with our in-house developed CALYPSO structure prediction method, we developed two acceleration schemes for structure prediction toward large systems, in which ML potential is pre-constructed to fully replace DFT calculations or trained in an on-the-fly manner from scratch during the structure searches The developed schemes have been applied to medium- and large-sized boron clusters, which are challenging cases for both construction of ML potentials and extensive structure searches Experimental structures of B36 and B40 clusters can be readily reproduced, and the putative global minimum structure for B84 cluster is proposed, where substantially less computational cost by several orders of magnitude is evident if compared with full DFT-based structure searches Our results demonstrate a viable route for structure prediction toward large systems via the combination of state-of-art structure prediction methods and ML techniques

Accelerating CALYPSO Structure Prediction by Data-driven Learning of Potential Energy Surface

The theoretical structure prediction method via quantum mechanical atomistic simulations such as density functional theory (DFT), based solely on chemical composition, has already become a routine tool to determine the structures of physical and chemical systems, e.g., solids and clusters. However, the application of DFT to more realistic simulations, to a large extent, is impeded because of the unfavorable scaling of the computational cost with respect to the system size. During recent years, the machine learning potential (MLP) method has been rapidly rising as an accurate and efficient tool for atomistic simulations. In this Perspective, we provide an introduction to the basic principles and advantages of the combination of structure prediction and MLP, as well as the challenges and opportunities associated with this promising approach.

Qunchao Tong

Papers

Accelerating CALYPSO structure prediction by data-driven learning of a potential energy surface

Accelerating CALYPSO Structure Prediction by Data-driven Learning of Potential Energy Surface

Combining Machine Learning Potential and Structure Prediction for Accelerated Materials Design and Discovery

X-ray diffraction data-assisted structure searches

The CALYPSO methodology for structure prediction