A latent-space extrapolation grade built into graph atomic cluster expansion foundation potentials

Yury Lysogorskiy; Anton Bochkarev; Ralf Drautz

Home/Archive/arXiv

arXiv · 2609.40060

A latent-space extrapolation grade built into graph atomic cluster expansion foundation potentials

Yury Lysogorskiy·Anton Bochkarev·Ralf Drautz

View original source ↗Read source document ↗Export metadata ↓

Abstract

Foundation machine-learning interatomic potentials cover broad configurational and chemical spaces, but their reliability can vary across the atomic environments encountered during a simulation. Here we introduce the calibrated Mahalanobis (CALM) extrapolation grade $γ$, a piecewise differentiable per-atom quantity integrated into GRACE foundation models and evaluated alongside energies and forces in a single model pass. We define $γ$ from nearest-cluster Mahalanobis distances in latent feature space, setting $γ=1$ from the training-distance distribution separately for each element and cluster. Controlled tests show that a normalized random projection of the invariant many-body basis detects structural and chemical extrapolation. On different foundation datasets, OMat24 and SMAX, $γ$ correlates with atomic force errors and separates structures with different error distributions. The CALM grade adds percent-level computational cost, and its spatial gradient guides uncertainty-biased data collection toward configurations with larger absolute force errors.

Explore related subjects

cond-mat.mtrl-sci

Keep this discovery

Explore connections, maps & timelines

BibTeX RIS

Yury Lysogorskiy, Anton Bochkarev, Ralf Drautz. 2026-09-30. A latent-space extrapolation grade built into graph atomic cluster expansion foundation potentials. https://arxiv.org/abs/2609.40060

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

KEEP EXPLORING

Related papers

arXiv2026

Surprisingly High Redundancy in Electronic Structure Data Across Materials Explained by Low Intrinsic Dimensionality

Machine learning (ML) models for electronic structure typically rely on large datasets generated by computationally expensive Kohn-Sham density functional theory calculations, as it is not known a priori which portions of the data are essential for accurate learning. Here, we reveal significant redundancies in electronic structure datasets across diverse material systems and attribute them to the low intrinsic dimensionality of the underlying data. We show that even random pruning can substantially reduce dataset size with minimal degradation in predictive accuracy. Moreover, a state-of-the-art coverage-based pruning strategy that samples data across all learning difficulties almost always preserves chemical accuracy and maintains model generalizability while using up to two orders of magnitude less data and reducing training time by a factor of three or more. We further demonstrate that the essential electronic structure information lies on a low-dimensional, non-linear manifold, providing a potential geometric explanation for the observed prunability. These observations are consistent with the predominance of local atomic environments in determining electronic properties, as suggested by nearsightedness arguments, and indicate that large-scale datasets may contain highly overlapping information. Our findings challenge the prevailing assumption that such extensive datasets are necessary for accurate ML-based electronic structure predictions and open a path toward identifying minimal, representative datasets for each material class.

cond-mat.mtrl-sci↗

arXiv2026

Large-gap quantum anomalous Hall insulator in two-dimensional pentagonal TiN$_8$ monolayer

Discovering the quantum anomalous Hall (QAH) effect in two-dimensional magnets is a central goal of topological condensed matter physics, yet candidate materials combining a sizable gap with a well-defined Chern number remain scarce. We report a new material platform that realizes emergent magnetism and nontrivial band topology within the largely overlooked family of two-dimensional MN$_8$ monolayers. Through first-principles calculations, we reveal that these systems exhibit out-of-plane magnetic ground states and nontrivial topological features driven by the localized $d$-orbitals of the embedded transition metals. Remarkably, we identify TiN$_8$ as a quantum anomalous Hall insulator characterized by a Chern number of $C=-1$ with a large topological gap of 223 meV. As a companion system, MoN$_8$ is predicted to be a rare high-Chern-number QAH insulator, with a Chern number of $C=2$, although its much smaller gap of 12 meV renders it more fragile. These findings establish the penta-MN$_8$ family as a versatile and promising platform to realize exotic topological quantum states. This work not only broadens the material landscape for magnetic topological insulators but also lays a solid theoretical foundation for the development of next-generation spintronic and quantum computing devices.

cond-mat.mtrl-sci↗

arXiv2026

Prediction of Magnetic Topological Materials Combining Spin and Magnetic Space Groups

Recent developments in spin space groups (SSGs) have led to a new classification of magnetic order and, with it, the emergence of altermagnetism. However, an efficient and routine approach that predicts electronic band topology in magnetic materials by SSGs, and especially the evolution of band topology from negligible spinorbit coupling (SOC) to finite SOC with the symmetry lowering from SSGs to magnetic space groups (MSGs) determined by magnetic-moment directions, has not yet been established. Here, we propose a scheme combining SSG and its MSG subgroups to diagnose band topology in collinear magnets using symmetry indicators of electronic band topology, established for all the 1,421 SSGs and 3,420 MSG subgroups. Specifically, the compatibility relations from SSG to the MSG subgroups can enable a complete topological classification for all possible magnetic-moment directions from the first-principles calculated numbers of occurrences of irreducible (co-)representations at high-symmetry points without SOC with respect to SSG. This scheme can be directly applied to collinear magnets whose magnetic structures have already been determined experimentally or theoretically. Applying it to 488 collinear magnets from MAGNDATA with experimentally determined magnetic structures, we identify 26 materials that are trivial by MSG but are expected to inherit topology protected by SSG symmetry. We showcase Dirac nodal lines and boundary states predicted by SSG but invisible to MSG in FePSe$_3$. Our work demonstrates the predictive power of the approach combining SSG and MSG over the approaches based on either group alone. The results of high-throughput calculations are expected to extend the magnetic topological materials pool, facilitating future experimental realizations of more magnetic topological materials.

cond-mat.mtrl-sci↗