Fast Kernel Methods for Generic Lipschitz Losses via $p$-Sparsified Sketches
This work addresses computational bottlenecks in kernel methods for machine learning practitioners, offering incremental improvements through sparsification and decomposition tricks.
The paper tackles the computational inefficiency of kernel methods by introducing sparsified sketches that maintain statistical accuracy while significantly reducing time and space requirements, with experiments showing empirical superiority over state-of-the-art sketching methods.
Kernel methods are learning algorithms that enjoy solid theoretical foundations while suffering from important computational limitations. Sketching, which consists in looking for solutions among a subspace of reduced dimension, is a well studied approach to alleviate these computational burdens. However, statistically-accurate sketches, such as the Gaussian one, usually contain few null entries, such that their application to kernel methods and their non-sparse Gram matrices remains slow in practice. In this paper, we show that sparsified Gaussian (and Rademacher) sketches still produce theoretically-valid approximations while allowing for important time and space savings thanks to an efficient \emph{decomposition trick}. To support our method, we derive excess risk bounds for both single and multiple output kernel problems, with generic Lipschitz losses, hereby providing new guarantees for a wide range of applications, from robust regression to multiple quantile regression. Our theoretical results are complemented with experiments showing the empirical superiority of our approach over SOTA sketching methods.