English

Block Basis Factorization for Scalable Kernel Matrix Evaluation

Machine Learning 2021-05-05 v4 Machine Learning Numerical Analysis Numerical Analysis

Abstract

Kernel methods are widespread in machine learning; however, they are limited by the quadratic complexity of the construction, application, and storage of kernel matrices. Low-rank matrix approximation algorithms are widely used to address this problem and reduce the arithmetic and storage cost. However, we observed that for some datasets with wide intra-class variability, the optimal kernel parameter for smaller classes yields a matrix that is less well approximated by low-rank methods. In this paper, we propose an efficient structured low-rank approximation method -- the Block Basis Factorization (BBF) -- and its fast construction algorithm to approximate radial basis function (RBF) kernel matrices. Our approach has linear memory cost and floating-point operations for many machine learning kernels. BBF works for a wide range of kernel bandwidth parameters and extends the domain of applicability of low-rank approximation methods significantly. Our empirical results demonstrate the stability and superiority over the state-of-art kernel approximation algorithms.

Keywords

Cite

@article{arxiv.1505.00398,
  title  = {Block Basis Factorization for Scalable Kernel Matrix Evaluation},
  author = {Ruoxi Wang and Yingzhou Li and Michael W. Mahoney and Eric Darve},
  journal= {arXiv preprint arXiv:1505.00398},
  year   = {2021}
}

Comments

16 pages, 5 figures

R2 v1 2026-06-22T09:27:09.378Z