TT-LSQR for tensor least squares problems and application to data mining
Lorenzo Piccinini, Valeria Simoncini · 2025
Read TT-LSQR for tensor least squares problems and application to data mining by Lorenzo Piccinini, Valeria Simoncini (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-2502-01293.txt.
Description
Lorenzo Piccinini and Valeria Simoncini adapt a classic solver to tensor problems in a 2025 paper. Tensors are grids of numbers with many axes, and their least squares problems ask for the tensor that best fits a target under a set of mixing matrices. Such problems grow huge fast, so the solver must stay light.
Their method, TT-LSQR, rebuilds the well-known LSQR algorithm around the tensor-train format, which stores a tensor as a chain of small pieces. A sketching step cuts the cost of the tall matrices involved. They apply the method to sorting a new document among filed ones by keyword, a task from information retrieval. The paper shows the tensor route handles the job at lower cost.
Catalog record
| CREATOR | Lorenzo Piccinini, Valeria Simoncini |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 54 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2502-01293