DeepAI AI Chat
Log In Sign Up

Near Input Sparsity Time Kernel Embeddings via Adaptive Sampling

by   David P. Woodruff, et al.

To accelerate kernel methods, we propose a near input sparsity time algorithm for sampling the high-dimensional feature space implicitly defined by a kernel transformation. Our main contribution is an importance sampling method for subsampling the feature space of a degree q tensoring of data points in almost input sparsity time, improving the recent oblivious sketching method of (Ahle et al., 2020) by a factor of q^5/2/ϵ^2. This leads to a subspace embedding for the polynomial kernel, as well as the Gaussian kernel, with a target dimension that is only linearly dependent on the statistical dimension of the kernel and in time which is only linearly dependent on the sparsity of the input dataset. We show how our subspace embedding bounds imply new statistical guarantees for kernel ridge regression. Furthermore, we empirically show that in large-scale regression tasks, our algorithm outperforms state-of-the-art kernel approximation methods.


page 1

page 2

page 3

page 4


Learning with Neural Tangent Kernels in Near Input Sparsity Time

The Neural Tangent Kernel (NTK) characterizes the behavior of infinitely...

Oblivious Sketching of High-Degree Polynomial Kernels

Kernel methods are fundamental tools in machine learning that allow dete...

Spectral Norm of Random Kernel Matrices with Applications to Privacy

Kernel methods are an extremely popular set of techniques used for many ...

ParK: Sound and Efficient Kernel Ridge Regression by Feature Space Partitions

We introduce ParK, a new large-scale solver for kernel ridge regression....

Leverage Score Sampling for Tensor Product Matrices in Input Sparsity Time

We give an input sparsity time sampling algorithm for spectrally approxi...

Distributed Adaptive Sampling for Kernel Matrix Approximation

Most kernel-based methods, such as kernel or Gaussian process regression...

Nyström Subspace Learning for Large-scale SVMs

As an implementation of the Nyström method, Nyström computational regula...