GPy-ABCD: A Configurable Automatic Bayesian Covariance Discovery Implementation

Thomas Fletcher; Alan Bundy; Kwabena Nuamah

GPy-ABCD: A Configurable Automatic Bayesian Covariance Discovery Implementation

Thomas Fletcher, Alan Bundy, Kwabena Nuamah

Research output: Contribution to conference › Poster › peer-review

Abstract / Description of output

Gaussian Processes (GPs) are a very flexible class of nonparametric models which are able to fit data with very few assumptions, namely just the type of correlation (kernel) the data is expected to display. Automatic Bayesian Covariance Discovery (ABCD)1 is an iterative modular Gaussian Process regression framework aimed at removing the requirement for even this initial correlation form assumption. GPy-ABCD2 is a new implementation of an ABCD system built for ease of use and configurability; it can produce short text descriptions of fit models, it uses a revised model-space search algorithm and it removes a search bias which was required in order to retain model explainability in the original system.

Original language	English
Number of pages	1
Publication status	Published - 23 Jul 2021
Event	8th ICML Workshop on Automated Machine Learning: Collocated with the Thirty-eighth International Conference on Machine Learning (ICML) - Virtual Duration: 23 Jul 2021 → 23 Jul 2021 https://sites.google.com/view/automl2021/home

Workshop

Workshop	8th ICML Workshop on Automated Machine Learning
Abbreviated title	AutoML
Period	23/07/21 → 23/07/21
Internet address	https://sites.google.com/view/automl2021/home

Access to Document

GPy-ABCD-poster_FLETCHER_DOA21062021_AFVAccepted author manuscript, 453 KBLicence: All Rights Reserved

1 Paper

GPy-ABCD: A Configurable Automatic Bayesian Covariance Discovery Implementation
Fletcher, T., Bundy, A. & Nuamah, K., 23 Jul 2021. 16 p.
Research output: Contribution to conference › Paper › peer-review

Open Access
File

Cite this

@conference{775186975ec14096a2511e689c509249,

title = "GPy-ABCD: A Configurable Automatic Bayesian Covariance Discovery Implementation",

abstract = "Gaussian Processes (GPs) are a very flexible class of nonparametric models which are able to fit data with very few assumptions, namely just the type of correlation (kernel) the data is expected to display. Automatic Bayesian Covariance Discovery (ABCD)1 is an iterative modular Gaussian Process regression framework aimed at removing the requirement for even this initial correlation form assumption. GPy-ABCD2 is a new implementation of an ABCD system built for ease of use and configurability; it can produce short text descriptions of fit models, it uses a revised model-space search algorithm and it removes a search bias which was required in order to retain model explainability in the original system.",

author = "Thomas Fletcher and Alan Bundy and Kwabena Nuamah",

year = "2021",

month = jul,

day = "23",

language = "English",

note = "8th ICML Workshop on Automated Machine Learning : Collocated with the Thirty-eighth International Conference on Machine Learning (ICML), AutoML ; Conference date: 23-07-2021 Through 23-07-2021",

url = "https://sites.google.com/view/automl2021/home",

}

TY - CONF

T1 - GPy-ABCD: A Configurable Automatic Bayesian Covariance Discovery Implementation

AU - Fletcher, Thomas

AU - Bundy, Alan

AU - Nuamah, Kwabena

PY - 2021/7/23

Y1 - 2021/7/23

N2 - Gaussian Processes (GPs) are a very flexible class of nonparametric models which are able to fit data with very few assumptions, namely just the type of correlation (kernel) the data is expected to display. Automatic Bayesian Covariance Discovery (ABCD)1 is an iterative modular Gaussian Process regression framework aimed at removing the requirement for even this initial correlation form assumption. GPy-ABCD2 is a new implementation of an ABCD system built for ease of use and configurability; it can produce short text descriptions of fit models, it uses a revised model-space search algorithm and it removes a search bias which was required in order to retain model explainability in the original system.

AB - Gaussian Processes (GPs) are a very flexible class of nonparametric models which are able to fit data with very few assumptions, namely just the type of correlation (kernel) the data is expected to display. Automatic Bayesian Covariance Discovery (ABCD)1 is an iterative modular Gaussian Process regression framework aimed at removing the requirement for even this initial correlation form assumption. GPy-ABCD2 is a new implementation of an ABCD system built for ease of use and configurability; it can produce short text descriptions of fit models, it uses a revised model-space search algorithm and it removes a search bias which was required in order to retain model explainability in the original system.

M3 - Poster

T2 - 8th ICML Workshop on Automated Machine Learning

Y2 - 23 July 2021 through 23 July 2021

ER -

GPy-ABCD: A Configurable Automatic Bayesian Covariance Discovery Implementation

Abstract / Description of output

Workshop

Access to Document

Fingerprint

Research output

GPy-ABCD: A Configurable Automatic Bayesian Covariance Discovery Implementation

Cite this