Search arXiv⌕ Search
Home/Archive/arXiv

arXiv · 2202.13019

Hierarchical symbolic regression for identifying key physical parameters correlated with bulk properties of perovskites

Lucas Foppa·Thomas A. R. Purcell·Sergey V. Levchenko·Matthias Scheffler·Luca M. Ghiringhelli
View original source ↗Read source document ↗Export metadata ↓

Abstract

Symbolic regression identifies key physical parameters describing materials properties by uncovering correlations as nonlinear analytical expressions. However, the pool of expressions grows rapidly with complexity, compromising its efficiency. We tackle this challenge by a hierarchical approach: identified expressions are used as input parameters for obtaining more complex expressions. Crucially, this framework can transfer knowledge among properties, highlighting physical relationships. We demonstrate this strategy by using the Sure-Independence-Screening-and-Sparsifying-Operator (SISSO) approach to identify expressions correlated with the lattice constant and cohesive energy, which are then used to model the bulk modulus of ABO3 perovskites.

Explore related subjects

cond-mat.mtrl-sciphysics.comp-ph

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

Lucas Foppa, Thomas A. R. Purcell, Sergey V. Levchenko, Matthias Scheffler, Luca M. Ghiringhelli. 2022-02-25. Hierarchical symbolic regression for identifying key physical parameters correlated with bulk properties of perovskites. https://doi.org/10.1103/physrevlett.129.055301

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related papers

arXiv2026

A hierarchy of thermodynamics learning frameworks for inelastic constitutive modeling

Recent advances in physics-augmented neural networks have enabled thermodynamically consistent data-driven constitutive modeling of complex inelastic materials. Most existing approaches, however, implicitly adopt a specific thermodynamic framework and embed structural assumptions such as normality, dual dissipation potentials, or other structure from manually constructed models directly into the learning architecture. Consequently, differences in predictive performance may arise not only from data or network design, but also from the underlying theoretical assumptions. In this work, we present a unified comparison of several thermodynamically consistent inelastic modeling frameworks from a machine learning perspective. We consider internal-variable formulations with dissipation potential, generalized standard materials, and metriplectic structures, and we analyze their structural assumptions, admissible dependencies, convexity requirements, and implications for dissipation and evolution. Each framework is implemented within a common neural potential architecture based on invariant representations and neural ordinary differential equations. This unified setting ensures that performance differences can be attributed to thermodynamic structure rather than architectural variation. The models are trained and evaluated on three representative inelastic datasets generated from high-fidelity representative volume element simulations: an elastoplastic alloy, a viscoelastic composite, and a rate-dependent crystal plasticity polycrystal. By isolating the role of thermodynamic structure, we assess how restrictions such as duality, normality, operator-based evolution, and convexity influence learnability, expressiveness, stability, and generalization.

cond-mat.mtrl-sci↗
arXiv2026

Berry Curvature Driven Transport in Silicon-Compatible Altermagnetic $α$-MnTe Thin Films

Integrating spin-dependent functionality with mainstream semiconductor technology is a central goal of modern spintronics, yet most candidate materials remain incompatible with silicon-based platforms. Here, we report the direct epitaxial integration of $α$-MnTe thin films on Si(111) via molecular beam epitaxy and demonstrate a robust anomalous Hall effect (AHE) in this silicon-compatible altermagnetic system. Despite the absence of net bulk magnetization, the films exhibit a pronounced hysteretic Hall response, providing transport evidence consistent with finite Berry curvature generated by symmetry breaking in the thin-film geometry. High-resolution structural and spectroscopic characterization confirms phase-pure, epitaxial growth with hexagonal NiAs-type symmetry, while magnetotransport measurements reveal correlated hysteresis in both transverse and longitudinal channels with systematic temperature evolution. First-principles calculations reveal substantial uncompensated Berry curvature arising from the spin-split band structure, consistent with altermagnetic symmetry and the origin of the observed Hall response. These results establish MnTe/Si(111) as a silicon-compatible altermagnetic platform and chart a concrete pathway for embedding Berry-phase-driven functionalities into scalable semiconductor device architectures.

cond-mat.mtrl-sci↗
arXiv2026

All-Optical Control of Interfacial Polarization in MoS$_2$/WSe$_2$ Heterobilayers

All-optical tuning of van der Waals heterostructures with coherent radiation offers a promising path toward ultrafast memory and optoelectronic devices. In the first-principles framework of real-time time-dependent density functional theory, we predict the induction of a persistent, long-lived out-of-plane polarization in MoS$_2$/WSe$_2$ heterobilayers, resonantly driven by intense ultrafast pulses. While weak fields preserve the intrinsic type-II band alignment, intermediate intensities trigger a four-fold enhancement of interlayer charge transfer. By analyzing the high-harmonic generation spectrum, we identify a transition from the perturbative to the strong-field regime inducing photoinduced interfacial polarity. We additionally show that lattice strain, ubiquitously present in heterobilayers, can be used as additional knob to adjust the resonant condition without compromising the permanent dipole induction. Our findings provide a theoretical blueprint for the all-optical manipulation of polar phases in low-dimensional heterostructures at the femtosecond scale.

cond-mat.mtrl-sci↗