Spectrum-Revealing Cholesky Factorization for Kernel Methods
This work provides a faster low-rank approximation for kernel methods, benefiting practitioners who need efficient solutions for large-scale machine learning problems.
The authors introduce spectrum-revealing Cholesky factorization for kernel matrix approximation, achieving comparable effectiveness to existing Cholesky-based methods while being significantly faster.
Kernel methods represent some of the most popular machine learning tools for data analysis. Since exact kernel methods can be prohibitively expensive for large problems, reliable low-rank matrix approximations and high-performance implementations have become indispensable for practical applications of kernel methods. In this work, we introduce spectrum-revealing Cholesky factorization, a reliable low-rank matrix factorization, for kernel matrix approximation. We also develop an efficient and effective randomized algorithm for computing this factorization. Our numerical experiments demonstrate that this algorithm is as effective as other Cholesky factorization based kernel methods on machine learning problems, but significantly faster.