Standing-wave-decomposition Gaussian process

Chi-Ken Lu, Scott Cheng-Hsin Yang, and Patrick Shafto
Phys. Rev. E 98, 032303 – Published 11 September 2018

Abstract

We propose a standing-wave-decomposition (SWD) approximation to Gaussian process (GP) regression. GP involves a costly matrix inversion operation, which limits applicability to large data analysis. For an input space that can be approximated by a grid and when correlations among data are short ranged, the kernel matrix inversion can be replaced by analytic diagonalization using the SWD. We show that this approach applies to uni- and multidimensional input data, extends to include longer-range correlations, and the grid can be in a latent space and used as inducing points. Through simulations, we show that our approximate method applied to the squared exponential kernel outperforms existing methods in predictive accuracy per unit time in the regime where data are plentiful. Our SWD-GP is recommended for regression analyses where there is a relatively large amount of data and/or there are constraints on computation time.

  • Figure
  • Figure
  • Figure
  • Figure
  • Figure
  • Figure
  • Figure
2 More
  • Received 24 May 2018
  • Revised 10 August 2018

DOI:https://doi.org/10.1103/PhysRevE.98.032303

©2018 American Physical Society

Physics Subject Headings (PhySH)

NetworksStatistical Physics & Thermodynamics

Authors & Affiliations

Chi-Ken Lu, Scott Cheng-Hsin Yang, and Patrick Shafto

  • Department of Mathematics and Computer Science, Rutgers University, Newark, New Jersey 07302, USA

Article Text (Subscription Required)

Click to Expand

References (Subscription Required)

Click to Expand
Issue

Vol. 98, Iss. 3 — September 2018

Reuse & Permissions
Access Options
CHORUS

Article Available via CHORUS

Download Accepted Manuscript
Author publication services for translation and copyediting assistance advertisement

Authorization Required


×
×

Images

×

Sign up to receive regular email alerts from Physical Review E

Log In

Cancel
×

Search


Article Lookup

Paste a citation or DOI

Enter a citation
×