Polarizable atomic multipoles for learning long-range electrostatics
This work addresses the challenge of modeling long-range electrostatics and polarization in MLIPs for ionic, polar, and interfacial systems, enabling prediction of polarization-sensitive observables without additional training data.
The authors introduce a semi-local framework using polarizable atomic multipoles to learn long-range electrostatics in machine learning interatomic potentials, achieving systematic improvements in potential energy surface accuracy across four benchmarks and recovering physically meaningful electrical responses such as Born effective charges and infrared spectra.
Long-range electrostatics and polarization remain central obstacles to extending machine learning interatomic potentials (MLIPs) to ionic, polar, and interfacial systems. Here, we introduce a semi-local framework for learning electrostatics from energies and forces using polarizable atomic multipoles. Local equivariant descriptors predict environment-dependent latent monopoles, dipoles, and quadrupoles, while residual non-local charge transfer and polarization are captured by non-self-consistent linear response in induced charges and dipoles. Across four diverse benchmarks and four short-range MLIP architectures, the multipole hierarchy and response terms systematically improve potential energy surface accuracy, with the largest gains in systems where long-range effects are essential. More importantly, the learned latent variables recover physically meaningful electrical responses: accurate Born effective charge tensors, emergent polarizabilities, infrared spectra in close agreement with experiments, and semi-quantitative Raman spectra for bulk water and hybrid MAPbI$_3$ perovskite. This systematically improvable, physically transparent framework enables MLIPs trained on standard energy and force labels to predict polarization-sensitive observables.