Fast Tensor Completion via Approximate Richardson Iteration
Mehrdad Ghadiri, Matthew Fahrbach, Yunbum Kook, Ali Jadbabaie · 2025
Read Fast Tensor Completion via Approximate Richardson Iteration by Mehrdad Ghadiri, Matthew Fahrbach, Yunbum Kook, Ali Jadbabaie (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-2502-09534.txt.
Description
Mehrdad Ghadiri, Matthew Fahrbach, Yunbum Kook, and Ali Jadbabaie published a paper in 2025 on filling in missing entries of multi-way data tables, called tensors. Recommendation lists and sensor grids often arrive with holes. Many fast filling methods lean on neat algebraic structure that the filling step itself destroys, leaving them hard to speed up.
The authors propose a lifting trick that lets structured solvers handle the messy filling problems as subroutines. An approximate iteration built on that trick comes with a proven convergence rate. On real data their method ran up to one hundred times faster than direct solvers for one standard decomposition. The gain opens large, sparse data sets to quick completion.
Catalog record
| CREATOR | Mehrdad Ghadiri, Matthew Fahrbach, Yunbum Kook, Ali Jadbabaie |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 74 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2502-09534