A Scalable Approach to Column-Based Low-Rank Matrix Approximation
Abstract
In this paper, we address the column-based low-rank matrix approximation problem using a novel parallel approach. Our approach is based on the divide-and-combine idea. We first perform column selection on submatrices of an original data matrix in parallel, and then combine the selected columns into the final output. Our approach enjoys a theoretical relative-error upper bound. In addition, our column-based low-rank approximation partitions data in a deterministic way and makes no assumptions about matrix coherence. Compared with other traditional methods, our approach is scalable on large-scale matrices. Finally, experiments on both simulated and real world data show that our approach is both efficient and effective.
Cite
Text
Pi et al. "A Scalable Approach to Column-Based Low-Rank Matrix Approximation." International Joint Conference on Artificial Intelligence, 2013.Markdown
[Pi et al. "A Scalable Approach to Column-Based Low-Rank Matrix Approximation." International Joint Conference on Artificial Intelligence, 2013.](https://mlanthology.org/ijcai/2013/pi2013ijcai-scalable/)BibTeX
@inproceedings{pi2013ijcai-scalable,
title = {{A Scalable Approach to Column-Based Low-Rank Matrix Approximation}},
author = {Pi, Yifan and Peng, Haoruo and Zhou, Shuchang and Zhang, Zhihua},
booktitle = {International Joint Conference on Artificial Intelligence},
year = {2013},
pages = {1600-1606},
url = {https://mlanthology.org/ijcai/2013/pi2013ijcai-scalable/}
}