Faster Kernel Matrix Algebra via Density Estimation

02/16/2021
by   Arturs Backurs, et al.
25

We study fast algorithms for computing fundamental properties of a positive semidefinite kernel matrix K ∈ℝ^n × n corresponding to n points x_1,…,x_n ∈ℝ^d. In particular, we consider estimating the sum of kernel matrix entries, along with its top eigenvalue and eigenvector. We show that the sum of matrix entries can be estimated to 1+ϵ relative error in time sublinear in n and linear in d for many popular kernels, including the Gaussian, exponential, and rational quadratic kernels. For these kernels, we also show that the top eigenvalue (and an approximate eigenvector) can be approximated to 1+ϵ relative error in time subquadratic in n and linear in d. Our algorithms represent significant advances in the best known runtimes for these problems. They leverage the positive definiteness of the kernel matrix, along with a recent line of work on efficient kernel density estimation.

READ FULL TEXT

page 1

page 2

page 3

page 4

10/15/2018

Eigenvalue Analysis via Kernel Density Estimation

In this paper, we propose an eigenvalue analysis -- of system dynamics m...
04/08/2021

Linear and Sublinear Time Spectral Density Estimation

We analyze the popular kernel polynomial method (KPM) for approximating ...
09/07/2009

Kernels for Measures Defined on the Gram Matrix of their Support

We present in this work a new family of kernels to compare positive meas...
10/06/2019

Kernel Density Estimation for Totally Positive Random Vectors

We study the estimation of the density of a totally positive random vect...
10/11/2020

A kernel-independent sum-of-Gaussians method by de la Vallée-Poussin sums

Approximation of interacting kernels by sum of Gaussians (SOG) is freque...
10/06/2019

Kernel Density Estimation for Totally Positive RandomVectors

We study the estimation of the density of a totally positive random vect...
06/08/2021

The Fast Kernel Transform

Kernel methods are a highly effective and widely used collection of mode...