collaborators

9 papers

physics.chem-ph2026

Projected Hessian Learning: Fast Curvature Supervision for Accurate Machine-Learning Interatomic Potentials

Austin Rodriguez, Justin S. Smith, Sakib Matin +3

The Hessian matrix (second derivatives) encodes far richer local curvature of the potential energy surface than energies and forces alone. However, training machine-learning intera…

physics.chem-ph2026

Knowledge Distillation of Noisy Force Labels for Improved Coarse-Grained Force Fields

Feranmi V. Olowookere, Sakib Matin, Aleksandra Pachalieva +2

Molecular dynamics simulations are an integral tool for studying the atomistic behavior of materials under diverse conditions. However, they can be computationally demanding in wal…

physics.comp-ph2025

Multi-fidelity learning for interatomic potentials: Low-level forces and high-level energies are all you need

Mitchell Messerly, Sakib Matin, Alice E. A. Allen +5

The promise of machine learning interatomic potentials (MLIPs) has led to an abundance of public quantum mechanical (QM) training datasets. The quality of an MLIP is directly limit…

physics.chem-ph2025

Reactive Chemistry at Unrestricted Coupled Cluster Level: High-throughput Calculations for Training Machine Learning Potentials

Alice E. A. Allen, Rui Li, Sakib Matin +8

Accurately modeling chemical reactions at the atomistic level requires high-level electronic structure theory due to the presence of unpaired electrons and the need to properly des…

cond-mat.str-el2025

Spin dynamics of triple-Q magnetic orderings in a triangular lattice: Implications for multi-Q orderings in general two-dimensional lattices

Pyeongjae Park, Woonghee Cho, Chaebin Kim +7

Multi-Q magnetic structures on two-dimensional (2D) lattices provide a key route to realizing topological physics in 2D magnetism. A major experimental challenge is to unambiguousl…

physics.chem-ph2025

Ensemble Knowledge Distillation for Machine Learning Interatomic Potentials

Sakib Matin, Emily Shinkle, Yulia Pimonova +5

The quality of machine learning interatomic potentials (MLIPs) strongly depends on the quantity of training data as well as the quantum chemistry (QC) level of theory used. Dataset…