Physics-Informed Weakly Supervised Learning For Interatomic Potentials

Machine learning plays an increasingly important role in computational chemistry and materials science, complementing expensive ab initio and firstprinciples methods. However, machine-learned interatomic potentials (MLIPs) often struggle with generalization and robustness, leading to unphysical energy and force predictions in atomistic simulations. To address this, we propose a physics-informed, weakly supervised training framework for MLIPs. Our method introduces two novel loss functions: one based on Taylor expansions of the potential energy and another enforcing conservative force constraints. This approach enhances accuracy, particularly in lowdata regimes, and reduces the reliance on large, expensive training datasets. Extensive experiments across benchmark datasets show up to 2× reductions in energy and force errors for multiple baseline models. Additionally, our method improves the stability of molecular dynamics simulations and facilitates effective fine-tuning of ML foundation models on sparse, high-accuracy ab initio data. Code and scripts to reproduce the experiments are available at https://github. com/nec-research/PICPS-ML4Sci .