Time-series Gaussian process regression based on toeplitz computation of O(N2) operations and O(N)-level storage

Y. Zhang, W.E. Leithead, D.J. Leith

Research output: Chapter in Book/Report/Conference proceedingConference contribution book

83 Citations (Scopus)

Abstract

Gaussian process (GP) regression is a Bayesian nonparametric model showing good performance in various applications. However, its hyperparameter-estimating procedure may contain numerous matrix manipulations of O(N3) arithmetic operations, in addition to the O(N2)-level storage. Motivated by handling the real-world large dataset of 24000 wind-turbine data, we propose in this paper an efficient and economical Toeplitz-computation scheme for time-series Gaussian process regression. The scheme is of O(N2) operations and O(N)-level memory requirement. Numerical experiments substantiate the effectiveness and possibility of using this Toeplitz computation for very large datasets regression (such as, containing 10000~100000 data points).
Original languageEnglish
Title of host publicationProceedings of the 44th IEEE Conference on Decision and Control, 2005 and 2005 European Control Conference. CDC-ECC '05
Subtitle of host publicationCDC-ECC '05
PublisherIEEE
Pages3711-3716
Number of pages6
ISBN (Print)0-7803-9567-0
DOIs
Publication statusPublished - Dec 2005
Event44th IEEE Conference on Decision and Control and European Control Conference - Seville, Spain
Duration: 12 Dec 200515 Dec 2005

Conference

Conference44th IEEE Conference on Decision and Control and European Control Conference
Abbreviated titleCDC-ECC
Country/TerritorySpain
CitySeville
Period12/12/0515/12/05

Keywords

  • acceleration
  • aerodynamics
  • bayesian methods
  • noise measurement
  • predictive models
  • velocity measurement

Fingerprint

Dive into the research topics of 'Time-series Gaussian process regression based on toeplitz computation of O(N2) operations and O(N)-level storage'. Together they form a unique fingerprint.

Cite this