Streaming Tensor Train Approximation
SIAM Journal on Scientific Computing · 31.10.2023 · 45 (5) · A2610-A2631
DOI: 10.1137/22m1515045 ↗Kurz gefasst
Fachartikel in der Zeitschrift SIAM Journal on Scientific Computing, online erschienen am 04.10.2023. Verfasst von Daniel Kressner und 2 weiteren Autor:innen. Die Arbeit wurde bisher 23-mal zitiert (Crossref) und nennt 21 Literaturangaben. Gefördert von: SNSF.
Abstract (Originalsprache)
Tensor trains are a versatile tool to compress and work with high-dimensional data and functions. In this work we introduce the streaming tensor train approximation (STTA), a new class of algorithms for approximating a given tensor [Formula: see text] in the tensor train format. STTA accesses [Formula: see text] exclusively via two-sided random sketches of the original data, making it streamable and easy to implement in parallel—unlike existing deterministic and randomized tensor train approximations. This property also allows STTA to conveniently leverage structures in [Formula: see text], such as sparsity and various low-rank tensor formats, as well as linear combinations thereof. When Gaussian random matrices are used for sketching, STTA is admissible for an analysis that builds and extends upon existing results on the generalized Nyström approximation for matrices. Our results show that STTA can be expected to attain a nearly optimal approximation error if the sizes of the sketches are suitably chosen. A range of numerical experiments illustrates the performance of STTA compared to existing deterministic and randomized approaches.
Förderung
- SNSF
Zitieren
APA
Kressner, D., Vandereycken, B., & Voorhaar, R. (2023). Streaming Tensor Train Approximation. SIAM Journal on Scientific Computing, 45(5), A2610-A2631. https://doi.org/10.1137/22m1515045
Vancouver
Kressner D, Vandereycken B, Voorhaar R. Streaming Tensor Train Approximation. SIAM Journal on Scientific Computing. 2023;45(5):A2610-A2631. doi:10.1137/22m1515045
BibTeX
@article{Kressner2023,
author = {Kressner, Daniel and Vandereycken, Bart and Voorhaar, Rik},
title = {Streaming Tensor Train Approximation},
journal = {SIAM Journal on Scientific Computing},
year = {2023},
volume = {45},
number = {5},
pages = {A2610--A2631},
doi = {10.1137/22m1515045}
}
EUVIMED