Non-Stationary Spectral Kernels

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

16 Citations (Scopus)


We propose non-stationary spectral kernels for Gaussian process regression. We propose to model the spectral density of a non-stationary kernel function as a mixture of input-dependent Gaussian process frequency density surfaces. We solve the generalised Fourier transform with such a model, and present a family of non-stationary and non-monotonic kernels that can learn input-dependent and potentially long-range, non-monotonic covariances between inputs. We derive efficient inference using model whitening and marginalized posterior, and show with case studies that these kernels are necessary when modelling even rather simple time series, image or geospatial data with non-stationary characteristics.
Original languageEnglish
Title of host publicationAdvances in Neural Information Processing Systems 30
Subtitle of host publicationProceedings of NIPS2017
PublisherCurran Associates, Inc.
Publication statusPublished - 2017
MoE publication typeA4 Article in a conference publication
EventNIPS Symposium on Interpretable Machine Learning - Long Beach, Los Angeles, United States
Duration: 4 Dec 20179 Dec 2017
Conference number: 31

Publication series

NameAdvances in Neural Information Processing Systems
PublisherCurran Associates
ISSN (Print)1049-5258


ConferenceNIPS Symposium on Interpretable Machine Learning
CountryUnited States
CityLos Angeles

Fingerprint Dive into the research topics of 'Non-Stationary Spectral Kernels'. Together they form a unique fingerprint.

Cite this