Gaussian process quantile regression using expectation propagation

Research output: Contribution to conferencePaper

View graph of relations Save citation

Authors

Research units

Abstract

Direct quantile regression involves estimating a given quantile of a response variable as a function of input variables. We present a new framework for direct quantile regression where a Gaussian process model is learned, minimising the expected tilted loss function. The integration required in learning is not analytically tractable so to speed up the learning we employ the Expectation Propagation algorithm. We describe how this work relates to other quantile regression methods and apply the method on both synthetic and real data sets. The method is shown to be competitive with state of the art methods whilst allowing for the leverage of the full Gaussian process probabilistic framework.

Request a copy

Request a copy

Details

Original languageEnglish
StatePublished - 27 Jun 2012
Event29th International Conference on Machine Learning - Ediburgh, United Kingdom

Conference

Conference29th International Conference on Machine Learning
Abbreviated titleICML 2012
CountryUnited Kingdom
CityEdiburgh
Period26/06/121/07/12

    Keywords

  • Gaussian process, probablisitc modelling, decision theory

Employable Graduates; Exploitable Research

Copy the text from this field...