Numerical Methods for Kernel Slicing
By: Nicolaj Rux, Johannes Hertrich, Sebastian Neumayer
Potential Business Impact:
Finds patterns faster in huge amounts of data.
Kernels are key in machine learning for modeling interactions. Unfortunately, brute-force computation of the related kernel sums scales quadratically with the number of samples. Recent Fourier-slicing methods lead to an improved linear complexity, provided that the kernel can be sliced and its Fourier coefficients are known. To obtain these coefficients, we view the slicing relation as an inverse problem and present two algorithms for their recovery. Extensive numerical experiments demonstrate the speed and accuracy of our methods.
Similar Papers
Fast kernel methods: Sobolev, physics-informed, and additive models
Machine Learning (Stat)
Makes big computer learning tasks much faster.
Data-Efficient Kernel Methods for Learning Differential Equations and Their Solution Operators: Algorithms and Error Analysis
Machine Learning (Stat)
Teaches computers to solve math problems faster.
Low-rank approximation of analytic kernels
Numerical Analysis
Makes computer math faster for science.