Gaussian Process Derivative at Uncertain Input for SE Kernel

Nghiem, Truong X. Gaussian Process Derivative at Uncertain Input for SE Kernel. Technical Report. UNSPECIFIED. (Unpublished)

Preview

Text (Technical report) gpderiv_uncertain.pdf - Draft Version Download (513kB) | Preview

Abstract

Given a Gaussian Process with a zero mean and a Squared Exponential (SE) kernel. We are interested in the exact mean and covariance of the predictive distribution of the latent function f and its gradient ∂f/∂x at an uncertain input x ∼ N(μ,Σ). This technical note develops the calculations of these quantities and documents an implementation of these calculations in a Matlab function called gppred_exactmoments_se.

Item Type:	Monograph (Technical Report)
Keywords:	Gaussian Processes, Uncertainty propagation
Subjects:	Q Science > QA Mathematics Q Science > QA Mathematics > QA75 Electronic computers. Computer science
NAU Depositing Author Academic Status:	Faculty/Staff
Department/Unit:	College of Engineering, Informatics, and Applied Sciences > School of Informatics, Computing, and Cyber Systems
Date Deposited:	26 Jun 2019 21:23
URI:	http://openknowledge.nau.edu/id/eprint/5501

Actions (login required)

IR Staff Record View

Downloads