Sparse Coresets for SVD on Infinite Streams
Abstract
In streaming Singular Value Decomposition (SVD), -dimensional rows of a possibly infinite matrix arrive sequentially as points in . An -coreset is a (much smaller) matrix whose sum of square distances of the rows to any hyperplane approximates that of the original matrix to a factor. Our main result is that we can maintain a -coreset while storing only rows. Known lower bounds of rows show that this is nearly optimal. Moreover, each row of our coreset is a weighted subset of the input rows. This is highly desirable since it: (1) preserves sparsity; (2) is easily interpretable; (3) avoids precision errors; (4) applies to problems with constraints on the input. Previous streaming results for SVD that return a subset of the input required storing rows where is the number of rows seen so far. Our algorithm, with storage independent of , is the first result that uses finite memory on infinite streams. We support our findings with experiments on the Wikipedia dataset benchmarked against state-of-the-art algorithms.
Keywords
Cite
@article{arxiv.2002.06296,
title = {Sparse Coresets for SVD on Infinite Streams},
author = {Vladimir Braverman and Dan Feldman and Harry Lang and Daniela Rus and Adiel Statman},
journal= {arXiv preprint arXiv:2002.06296},
year = {2020}
}