Scalable Gaussian process modeling of parametrized spatio-temporal fields Journal: Journal of Computational Physics
Srinath Dama, Prasanth B. Nair · 2026
Read Scalable Gaussian process modeling of parametrized spatio-temporal fields Journal: Journal of Computational Physics by Srinath Dama, Prasanth B. Nair (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-2603-00290.txt.
Description
Dama and Nair introduce a Gaussian process framework with deep product kernels for fields that vary across space and time. The framework learns a continuous picture of such fields, so it can predict at any point regardless of resolution. Kronecker algebra makes training cheap, with cost growing almost linearly in grid size. A key strength is that posterior variance costs essentially the same as the posterior mean, enabling uncertainty estimates. Tests on benchmarks match operator-learning methods, and on the unsteady Burgers equation the method beats reduced-order models.
Catalog record
| CREATOR | Srinath Dama, Prasanth B. Nair |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 103863 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2603.00290