Parallel Approximate Matrix Factorization for Kernel Methods

Kaihua Zhu, Hang Cui, Hongjie Bai, Jian Li, Zhihuan Qiu, Hao Wang, Hui Xu, Edward Y. ChangICME 2007 · OpenReview

Abstract

The kernel methods play a pivotal role in machine learning algorithms. Unfortunately, working with the kernel methods must deal with an n times n kernel matrix, which is memory intensive. In this paper, we present a parallel, approximate matrix factorization algorithm, which loads only essential data to individual processors to enable parallel processing of data. Our method reduces space requirement for the kernel matrix from O(n <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> ) to O(np/m), where n is the amount of data, p the reduced matrix dimension (p << n), and m the number of processors.
BibTeX
@inproceedings{zhu2007parallel,
  author = {Kaihua Zhu and Hang Cui and Hongjie Bai and Jian Li and Zhihuan Qiu and Hao Wang and Hui Xu and Edward Y. Chang},
  title = {{Parallel Approximate Matrix Factorization for Kernel Methods}},
  booktitle = {ICME},
  year = {2007},
  url = {https://openreview.net/forum?id=nmHcJ8DmcSh},
}

Loading related content...

Figure