Atom-Resolved Machine Learning of Dielectric and Piezoelectric Response
Abstract
Predicting dielectric and piezoelectric responses in structurally complex materials requires large simulation cells, for which density-functional perturbation theory (DFPT) calculations scale as 𝒪(N4) and become computationally prohibitive. However, machine-learning approaches have remained challenging, limited either by the large amounts of expensive DFPT data required for direct regression or by the cubic cost of reconstruction. We here propose a machine-learning framework that learns the ionic response as an atom-indexed field: DART (Direct Atom-Resolved response-Tensor learning) learns these fields directly and achieves high accuracy from small training sets, whereas LARS (Linear-scaling Atom-Resolved Response Solver) reconstructs them from learned microscopic ingredients through sparse linear solves and requires no DFPT labels for the ionic dielectric or piezoelectric tensors. Using DART, we extrapolate to 314 stackings of AlN/ScN superlattices absent from training and identify a high-response polar candidate, for which DFPT gives a laterally clamped piezoelectric strain coefficient d33,f=14.511 pC/N and kt2=20.46%, exceeding the ordered 1AlN/1ScN reference by 63% and 66%.
Related
- Lead-free piezoelectric perovskites for arterial-pulse e-skin: from configurational complexity to equivariant machine-learning potentials
- Dielectric Response of Short and Long Range Models of Nonuniform Liquids and Implications for Machine Learned Interatomic Potentials
- ZStar: A unified toolkit for polarization, Born effective charges, dielectric and piezoelectric responses, and infrared and Raman spectra
- Reliability of Fisher-information bounds for dielectric metalenses: a site-resolved RCWA audit
- Linking Three-Dimensional Morphology to Piezoelectric Response in P(VDF-TrFE)
BibTeX
@misc{liu2026atomresolved,
title = {{Atom-Resolved Machine Learning of Dielectric and Piezoelectric Response}},
author = {Jinyu Liu and Yingwei Chen and Liyang Ma and Hongyu Yu and Hongjun Xiang},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2609.29978},
eprint = {2609.29978},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2609.29978},
}
RIS
TY - UNPB
TI - Atom-Resolved Machine Learning of Dielectric and Piezoelectric Response
AU - Jinyu Liu
AU - Yingwei Chen
AU - Liyang Ma
AU - Hongyu Yu
AU - Hongjun Xiang
PY - 2026
DA - 2026/09/24
PB - arXiv
DO - 10.48550/arxiv.2609.29978
UR - https://arxiv.org/abs/2609.29978
AB - Predicting dielectric and piezoelectric responses in structurally complex materials requires large simulation cells, for which density-functional perturbation theory (DFPT) calculations scale as 𝒪(N^4) and become computationally prohibitive. However, machine-learning approaches have remained challenging, limited either by the large amounts of expensive DFPT data required for direct regression or by the cubic cost of reconstruction. We here propose a machine-learning framework that learns the ionic response as an atom-indexed field: DART (Direct Atom-Resolved response-Tensor learning) learns these fields directly and achieves high accuracy from small training sets, whereas LARS (Linear-scaling Atom-Resolved Response Solver) reconstructs them from learned microscopic ingredients through sparse linear solves and requires no DFPT labels for the ionic dielectric or piezoelectric tensors. Using DART, we extrapolate to 314 stackings of AlN/ScN superlattices absent from training and identify a high-response polar candidate, for which DFPT gives a laterally clamped piezoelectric strain coefficient d_33,f=14.511 pC/N and k_t^2=20.46%, exceeding the ordered 1AlN/1ScN reference by 63% and 66%.
ER -