Gaussian Process Kernels for Pattern Discovery and Extrapolation

02/18/2013
by   Andrew Gordon Wilson, et al.
0

Gaussian processes are rich distributions over functions, which provide a Bayesian nonparametric approach to smoothing and interpolation. We introduce simple closed form kernels that can be used with Gaussian processes to discover patterns and enable extrapolation. These kernels are derived by modelling a spectral density -- the Fourier transform of a kernel -- with a Gaussian mixture. The proposed kernels support a broad class of stationary covariances, but Gaussian process inference remains simple and analytic. We demonstrate the proposed kernels by discovering patterns and performing long range extrapolation on synthetic examples, as well as atmospheric CO2 trends and airline passenger data. We also show that we can reconstruct standard covariances within our framework.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
11/27/2018

Neural Non-Stationary Spectral Kernel

Standard kernels such as Matérn or RBF kernels only encode simple monoto...
research
10/26/2015

The Human Kernel

Bayesian nonparametric models, such as Gaussian processes, provide a com...
research
11/15/2017

Spatial Mapping with Gaussian Processes and Nonstationary Fourier Features

The use of covariance kernels is ubiquitous in the field of spatial stat...
research
06/07/2015

String Gaussian Process Kernels

We introduce a new class of nonstationary kernels, which we derive as co...
research
02/02/2018

Scalable Lévy Process Priors for Spectral Kernel Learning

Gaussian processes are rich distributions over functions, with generaliz...
research
10/29/2019

Function-Space Distributions over Kernels

Gaussian processes are flexible function approximators, with inductive b...
research
10/09/2020

Sparse Spectrum Warped Input Measures for Nonstationary Kernel Learning

We establish a general form of explicit, input-dependent, measure-valued...

Please sign up or login with your details

Forgot password? Click here to reset