This item is only available for download by members of the University of Illinois community. Students, faculty, and staff at the U of I may log in with your NetID and password to view the item. If you are trying to access an Illinois-restricted dissertation or thesis, you can request a copy through your library's Inter-Library Loan office or purchase a copy directly from ProQuest.
Machine Learning is a constantly evolving field with main purpose to allow users to fit a model based on data and receive recommendations or predictions based on this model. In this thesis, we will specifically look at regression, which is a type of supervised learning where the output of the model is continuous.
A useful framework that can be used in regression tasks is the framework of Gaussian Processes (GPs). An important function characterizing the statistical properties of Gaussian Processes is their covariance function, which corresponds to a measure of similarity between datapoints. We will investigate the main motivation of using Gaussian Processes in regression tasks as compared to classical techniques, such as linear regression. We will also look at important variants of Gaussian Processes called Additive Gaussian Processes, which are inspired by Generalized Additive Models. To implement Additive Gaussian Processes, we will investigate how their covariance functions affect predictions and how summations of many covariance functions can improve predictions.