arXiv subjects

Mao Su

Publications and source records attributed to Mao Su.

23 records · Page 2Linked to original sources

arXiv2024

Geometry-enhanced Pre-training on Interatomic Potentials

Machine learning interatomic potentials (MLIPs) enables molecular dynamics (MD) simulations with ab initio accuracy and has been applied to various fields of physical science. However, the performance and transferability of MLIPs are limited by insufficient labeled training data, which require expensive ab initio calculations to obtain the labels, especially for complex molecular systems. To address this challenge, we design a novel geometric structure learning paradigm that consists of two stages. We first generate a large quantity of 3D configurations of target molecular system with classical molecular dynamics simulations. Then, we propose geometry-enhanced self-supervised learning consisting of masking, denoising, and contrastive learning to better capture the topology and 3D geometric information from the unlabeled 3D configurations. We evaluate our method on various benchmarks ranging from small molecule datasets to complex periodic molecular systems with more types of elements. The experimental results show that the proposed pre-training method can greatly enhance the accuracy of MLIPs with few extra computational costs and works well with different invariant or equivariant graph neural network architectures. Our method improves the generalization capability of MLIPs and helps to realize accurate MD simulations for complex molecular systems.

physics.chem-ph↗

arXiv2023

Transferable E(3) equivariant parameterization for Hamiltonian of molecules and solids

Using the message-passing mechanism in machine learning (ML) instead of self-consistent iterations to directly build the mapping from structures to electronic Hamiltonian matrices will greatly improve the efficiency of density functional theory (DFT) calculations. In this work, we proposed a general analytic Hamiltonian representation in an E(3) equivariant framework, which can fit the ab initio Hamiltonian of molecules and solids by a complete data-driven method and are equivariant under rotation, space inversion, and time reversal operations. Our model reached state-of-the-art precision in the benchmark test and accurately predicted the electronic Hamiltonian matrices and related properties of various periodic and aperiodic systems, showing high transferability and generalization ability. This framework provides a general transferable model that can be used to accelerate the electronic structure calculations on different large systems with the same network weights trained on small structures.

physics.comp-ph↗

arXiv2023

Efficient determination of the Hamiltonian and electronic properties using graph neural network with complete local coordinates

Despite the successes of machine learning methods in physical sciences, prediction of the Hamiltonian, and thus electronic properties, is still unsatisfactory. Here, based on graph neural network architecture, we present an extendable neural network model to determine the Hamiltonian from ab initio data, with only local atomic structures as inputs. Rotational equivariance of the Hamiltonian is achieved by our complete local coordinates. The local coordinates information, encoded using the convolutional neural network and designed to preserve Hermitian symmetry, is used to map hopping parameters onto local structures. We demonstrate the performance of our model using graphene and SiGe random alloys as examples. We show that our neural network model, although trained using small-size systems, can predict the Hamiltonian, as well as electronic properties such as band structures and densities of states (DOS) for large-size systems within the ab initio accuracy, justifying its extensibility. In combination with the high efficiency of our model, which takes only seconds to get the Hamiltonian of a 1728-atom system, present work provides a general framework to predict electronic properties efficiently and accurately, which provides new insights into computational physics and will accelerate the research for large-scale materials.

cond-mat.mtrl-sci↗

arXiv2020

A Brief Review of Continuous Models for Ionic Solutions: the Poisson-Boltzmann and Related Theories

The Poisson-Boltzmann (PB) theory is one of the most important theoretical models describing charged systems continuously. However, it suffers from neglecting ion correlations, which hinders its applicability to more general charged systems other than extremely dilute ones. Therefore, some modified versions of the PB theory are developed to effectively include ion correlations. Focused on their applications to ionic solutions, the original PB theory and its variances, including the field-theoretic approach, the correlation-enhanced PB model, the Outhwaite-Bhuiyan modified PB theory and the mean field theories, are briefly reviewed in this paper with the diagnosis of their advantages and limitations.

cond-mat.stat-mech↗

arXiv2019

Poisson-Boltzmann theory with non-linear ion correlations

The Poisson-Boltzmann (PB) theory is widely used to depict ionic systems. As a mean-field theory, the PB theory neglects the correlation effect in the ionic atmosphere and leads to deviations from experimental results as the concentration or charge valance increases. A modified PB theory including ion correlation effect while retaining its simplicity is critical for many important applications in which ion correlation effect can be significant. In this paper, we present a new model to incorporate ion correlations into the original PB equation by utilizing the Green's function with a non-linear form of the self-energy, which is different from the linear self-energy equation obtained by the Field-Theoretic (FT) approach. Both equations are solved numerically and compared with our molecular dynamics (MD) simulation. The co-ion distribution calculated by the FT approach deviates significantly from the MD simulation, while our results for both counter-ion and co-ion distributions are justified by the MD simulation.

cond-mat.soft↗