Don’t Get Your Kroneckers in a Twist: Gaussian Processes on High-Dimensional Incomplete Grids
Mads Greisen Højlund, August Smart Lykke-Møller, Henry Moss, Ove Christiansen · 2026
Read Don’t Get Your Kroneckers in a Twist: Gaussian Processes on High-Dimensional Incomplete Grids by Mads Greisen Højlund, August Smart Lykke-Møller, Henry Moss, Ove Christiansen (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2605-08036.txt.
Description
Mads Greisen Højlund, August Smart Lykke-Møller, Henry Moss, and Ove Christiansen make Gaussian process regression exact and fast on very large, high-dimensional data. Standard methods drown in the size of the kernel matrix, and data often arrive on incomplete grids that break older fast tricks.
Their method, CUTS-GPR, pairs an additive kernel with the incomplete grid so the kernel matrix gains a Kronecker-like structure with fast matrix-vector products. Benchmarks reach billions of points and thousands of dimensions. Full runs, including hyperparameter tuning, handled about four and a half million points in five hundred dimensions, plus ten real potential energy surfaces from chemistry.
Catalog record
| CREATOR | Mads Greisen Højlund, August Smart Lykke-Møller, Henry Moss, Ove Christiansen |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 160 KB |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2605.08036