GPflow: A Gaussian Process Library using TensorFlow

Alexander G.De G. Matthews; Mark Van Der Wilk; Tom Nickson; Keisuke Fujii; Alexis Boukouvalas; Pablo León-Villagrá; Zoubin Ghahramani; James Hensman

GPflow: A Gaussian Process Library using TensorFlow

Alexander G.De G. Matthews, Mark Van Der Wilk, Tom Nickson, Keisuke Fujii, Alexis Boukouvalas, Pablo León-Villagrá, Zoubin Ghahramani, James Hensman

Diagnostics and Control

Research output: Contribution to journal › Article › peer-review

520 Scopus citations

Abstract

GPflow is a Gaussian process library that uses TensorFlow for its core computations and Python for its front end.1 The distinguishing features of GPflow are that it uses variational inference as the primary approximation method, provides concise code through the use of automatic differentiation, has been engineered with a particular emphasis on software testing and is able to exploit GPU hardware.

Original language	English
Journal	Journal of Machine Learning Research
Volume	18
State	Published - Apr 1 2017

Funding

We acknowledge contributions from Valentine Svensson, Dan Marthaler, David J. Harris, Rasmus Munk Larsen and Eugene Brevdo. We acknowledge EPSRC grants EP/I036575/1, EP/N014162/1 and EP/N510129/1. James Hensman was supported by an MRC fellowship.

Cite this

@article{e2948cc15a7148b8af40220453b1de07,

title = "GPflow: A Gaussian Process Library using TensorFlow",

abstract = "GPflow is a Gaussian process library that uses TensorFlow for its core computations and Python for its front end.1 The distinguishing features of GPflow are that it uses variational inference as the primary approximation method, provides concise code through the use of automatic differentiation, has been engineered with a particular emphasis on software testing and is able to exploit GPU hardware.",

author = "Matthews, \{Alexander G.De G.\} and \{Van Der Wilk\}, Mark and Tom Nickson and Keisuke Fujii and Alexis Boukouvalas and Pablo Le{\'o}n-Villagr{\'a} and Zoubin Ghahramani and James Hensman",

note = "Publisher Copyright: {\textcopyright} 2017 Alexander G. de G. Matthews, Mark van der Wilk, Tom Nickson, Keisuke Fujii, Alexis Boukouvalas, Pablo Le{\'o}n-Villagr{\'a}, Zoubin Ghahramani, and James Hensman.",

year = "2017",

month = apr,

day = "1",

language = "English",

volume = "18",

journal = "Journal of Machine Learning Research",

issn = "1532-4435",

publisher = "Journal of Machine Learning Research",

}

TY - JOUR

T1 - GPflow

T2 - A Gaussian Process Library using TensorFlow

AU - Matthews, Alexander G.De G.

AU - Van Der Wilk, Mark

AU - Nickson, Tom

AU - Fujii, Keisuke

AU - Boukouvalas, Alexis

AU - León-Villagrá, Pablo

AU - Ghahramani, Zoubin

AU - Hensman, James

PY - 2017/4/1

Y1 - 2017/4/1

N2 - GPflow is a Gaussian process library that uses TensorFlow for its core computations and Python for its front end.1 The distinguishing features of GPflow are that it uses variational inference as the primary approximation method, provides concise code through the use of automatic differentiation, has been engineered with a particular emphasis on software testing and is able to exploit GPU hardware.

AB - GPflow is a Gaussian process library that uses TensorFlow for its core computations and Python for its front end.1 The distinguishing features of GPflow are that it uses variational inference as the primary approximation method, provides concise code through the use of automatic differentiation, has been engineered with a particular emphasis on software testing and is able to exploit GPU hardware.

UR - https://www.scopus.com/pages/publications/85020889788

M3 - Article

AN - SCOPUS:85020889788

SN - 1532-4435

VL - 18

JO - Journal of Machine Learning Research

JF - Journal of Machine Learning Research

ER -

GPflow: A Gaussian Process Library using TensorFlow

Abstract

Funding

Other files and links

Fingerprint

Cite this