Adding constrained discontinuities to Gaussian process models of wind fields

Dan Cornford; Ian T. Nabney; Christopher K. I. Williams; Michael S. Kearns; Sara A. Solla; David A. Cohn

Adding constrained discontinuities to Gaussian process models of wind fields

Dan Cornford, Ian T. Nabney, Christopher K. I. Williams, Michael S. Kearns (Editor), Sara A. Solla (Editor), David A. Cohn (Editor)

Computer Science Research Group

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

Abstract

Gaussian Processes provide good prior models for spatial data, but can be too smooth. In many physical situations there are discontinuities along bounding surfaces, for example fronts in near-surface wind fields. We describe a modelling method for such a constrained discontinuity and demonstrate how to infer the model parameters in wind fields with MCMC sampling.

Original language	English
Pages (from-to)	861-867
Number of pages	7
Journal	Advances in Neural Information Processing Systems
Volume	11
Publication status	Published - 1999
Event	Advances in Neural Information Processing Systems 1999 - Dublin, Ireland Duration: 3 Aug 1999 → 6 Aug 1999

Bibliographical note

Copyright of the Massachusetts Institute of Technology Press (MIT Press) Available also on Google Books

Keywords

Gaussian processes
spatial data
discontinuities
near-surface

Access to Document

0861.pdf
Copyright of the Massachusetts Institute of Technology Press (MIT Press) Available also on Google Books
Final published version, 839 KB

Cite this

@article{8fa05f55c9894046a7db9b68bdc12986,

title = "Adding constrained discontinuities to Gaussian process models of wind fields",

abstract = "Gaussian Processes provide good prior models for spatial data, but can be too smooth. In many physical situations there are discontinuities along bounding surfaces, for example fronts in near-surface wind fields. We describe a modelling method for such a constrained discontinuity and demonstrate how to infer the model parameters in wind fields with MCMC sampling.",

keywords = "Gaussian processes, spatial data, discontinuities, near-surface",

author = "Dan Cornford and Nabney, {Ian T.} and Williams, {Christopher K. I.} and Kearns, {Michael S.} and Solla, {Sara A.} and Cohn, {David A.}",

note = "Copyright of the Massachusetts Institute of Technology Press (MIT Press) Available also on Google Books; Advances in Neural Information Processing Systems 1999 ; Conference date: 03-08-1999 Through 06-08-1999",

year = "1999",

language = "English",

volume = "11",

pages = "861--867",

journal = "Advances in Neural Information Processing Systems",

issn = "1049-5258",

publisher = "MIT Press Journals",

}

TY - JOUR

T1 - Adding constrained discontinuities to Gaussian process models of wind fields

AU - Cornford, Dan

AU - Nabney, Ian T.

AU - Williams, Christopher K. I.

A2 - Kearns, Michael S.

A2 - Solla, Sara A.

A2 - Cohn, David A.

N1 - Copyright of the Massachusetts Institute of Technology Press (MIT Press) Available also on Google Books

PY - 1999

Y1 - 1999

N2 - Gaussian Processes provide good prior models for spatial data, but can be too smooth. In many physical situations there are discontinuities along bounding surfaces, for example fronts in near-surface wind fields. We describe a modelling method for such a constrained discontinuity and demonstrate how to infer the model parameters in wind fields with MCMC sampling.

AB - Gaussian Processes provide good prior models for spatial data, but can be too smooth. In many physical situations there are discontinuities along bounding surfaces, for example fronts in near-surface wind fields. We describe a modelling method for such a constrained discontinuity and demonstrate how to infer the model parameters in wind fields with MCMC sampling.

KW - Gaussian processes

KW - spatial data

KW - discontinuities

KW - near-surface

UR - http://mitpress.mit.edu/catalog/item/default.asp?ttype=2&tid=4746

M3 - Article

SN - 1049-5258

VL - 11

SP - 861

EP - 867

JO - Advances in Neural Information Processing Systems

JF - Advances in Neural Information Processing Systems

T2 - Advances in Neural Information Processing Systems 1999

Y2 - 3 August 1999 through 6 August 1999

ER -

Adding constrained discontinuities to Gaussian process models of wind fields

Abstract

Bibliographical note

Keywords

Access to Document

Other files and links

Fingerprint

Cite this