Research Overview

Our lab leverages machine learning and high-performance computing simulations to accelerate the discovery and understanding of new materials. We focus on the following key research thrusts:

Machine Learning and Electronic Structure Methods

We develop the Materials Learning Algorithms (MALA), a physics-informed machine learning framework designed to accelerate conventional density functional theory simulations. MALA employs neural network architectures to accurately predict electronic structures across different parameter spaces. Our scalable approach overcomes the limitations of conventional density functional theory simulations, paving the way for electronic structure calculations at unprecedented length and time scales.

Materials Learning Algorithms (MALA): Scalable Machine Learning for Electronic Structure Calculations in Large-Scale Atomistic Simulations MALA highlight

Predicting Electronic Structures at Any Length Scale with Machine Learning Predicting electronic structures highlight

Deep Dive into Machine Learning Density Functional Theory for Materials Science and Chemistry Deep dive into ML for DFT highlight

Key Areas: Neural Networks, Electronic Structure Theory, Density Functional Theory, Atomistic Simulations

Atomistic Molecular Spin Dynamics

We use a combination of first-principles calculations and machine learning models to generate interatomic potentials for high-performance molecular-spin dynamics simulations. This allows us to simulate ionic and spin dynamics simultaneously, enabling detailed analyses of structural stability, transport phenomena, and magneto-structural phase transitions. This approach shows promise in advancing next-generation magnetic materials and ultrafast magnetic storage technologies.

Probing Iron in Earth’s Core with Molecular-Spin Dynamics Molecular-spin dynamics highlight

Data-Driven Magneto-Elastic Predictions with Scalable Classical Spin-Lattice Dynamics Molecular-spin dynamics highlight

Key Areas: Machine Learning Interatomic Potentials, Atomistic Simulations, Molecular Dynamics, Spin Dynamics, Magnetic Materials

Explorative Artificial Intelligence for Materials Modeling

We leverage state-of-the-art machine learning techniques to advance and automate first-principles simulations, paving the way for rapid and targeted materials discovery. We employ physics-informed neural networks for inverting fundamental quantum mechanical equations, neural operators for modeling electron dynamics, and generative models for materials discovery.

Bridging the Gap in Electronic Structure Calculations via Machine Learning Bridging the Gap in Electronic Structure Calculations via Machine Learning highlight

Inverting the Kohn–Sham Equations with Physics-Informed Machine Learning Inverting the Kohn–Sham Equations with Physics-Informed Machine Learning highlight

Key Areas: Physics-Informed Neural Networks, Neural Operators, Time-Dependent Density Functional Theory, Exchange-Correlation Functionals, AI for Materials Discovery

More Research Highlights