Bayesian Gaussian processes for sequential prediction, optimisation and quadrature

Thesis

We develop a family of Bayesian algorithms built around Gaussian processes for various problems posed by sensor networks. We firstly introduce an iterative Gaussian process for multi-sensor inference problems, and show how our algorithm is able to cope with data that may be noisy, missing, delayed and/or correlated. Our algorithm can also effectively manage data that features changepoints, such as sensor faults. Extensions to our algorithm allow us to tackle some of the decision problems f...

Expand abstract

Authors: Osborne, M; Michael Alan Osborne

• Publication date: 2010
• Type of award: DPhil
• Level of award: Doctoral
• Awarding institution: Oxford University, UK
• Language: English
• Keywords: changepoint detection; observation selection; global optimisation; Bayesian inference; Bayesian quadrature; Gaussian processes; sensor networks
• Subjects: Information engineering; Sensors; Applications and algorithms; Probability theory and stochastic processes