Learning the Geometry of Materials Space with Trained Neural Tangent Kernels
Overview & Synopsis
Neural networks trained to predict materials properties learn representations whose physical content can be difficult to extract. We use the neural tangent kernel (NTK), evaluated at trained parameters, to recover the geometry learned by a network that predicts superconducting transition temperatures from elemental composition alone. Kernel ridge regression with the trained NTK closely reproduces the network's predictive performance and substantially outperforms regression with the initial kernel. Spectral analysis and multidimensional scaling reveal a geometry dominated by a small number of directions. Clustering and local neighborhood analysis organize compounds into recognizable superconducting families and connect materials with related chemical and correlated-electron properties despite substantial differences in $T_c$. These results establish trained tangent kernels as practical probes of learned materials representations, connecting property prediction with the exploration of materials space.
arxiv.org)
← / → for pages • + / - for zoom • 0 to fit width
Powered by Mozilla PDF.js v3.11
Document Information
| Document Title | Learning the Geometry of Materials Space with Trained Neural Tangent Kernels |
|---|---|
| Author / Organization | Daniel Kaplan, Gabriel Kotliar, Anirvan M Sengupta |
| Subject & Discipline | cond-mat.mtrl-sci |
| Categories | Science & Mathematics |
| Publication Year | 2026 |
| License / Rights | arXiv Open Access / Creative Commons |
| Resource Verification | Listed PDF (Verification Pending) |
| Original Host Website |
arXiv
(arxiv.org)
|
| File Format | application/pdf (Portable Document Format) |
| Language | EN |
| Catalog ID | DOC-271428 |
| Original Source Link | Visit Source Page on arxiv.org » |