Scalable Signature Kernel Computations via Local Neumann Series Expansions
Matthew Tamayo-Rios, Alexander Schell, Rima Alaifari · 2025
Read Scalable Signature Kernel Computations via Local Neumann Series Expansions by Matthew Tamayo-Rios, Alexander Schell, Rima Alaifari (2025) 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-2502-20392.txt.
Description
Matthew Tamayo-Rios, Alexander Schell, and Rima Alaifari speed up a kernel method for long data sequences, in work from 2025. The signature kernel scores how alike two sequences are and works well on rich, high-dimensional data. Its cost grows steeply with sequence length, though.
The kernel solves a classic boundary equation on a square domain. Their method cuts that square into tiles and solves each tile with a short local power series. Solutions then pass their edge values to the next tiles in a fixed order. Truncation adapts to local difficulty, trading cost against accuracy in a controlled way. Gains open the kernel to much longer records, such as finance and health series.
Catalog record
| CREATOR | Matthew Tamayo-Rios, Alexander Schell, Rima Alaifari |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 111 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2502-20392