Differential Equations for the Construction of Novel Kernels in Machine Learning Skip to main content
Utah's Foremost Platform for Undergraduate Research Presentation
2013 Abstracts

Differential Equations for the Construction of Novel Kernels in Machine Learning

Alexandra Pasi, Brigham Young University

Mathematics

Significant research in Machine Learning has been directed at the application and implementation of kernel-based learning methods. However, few studies have focused on the problem of kernel construction. This paper introduces a novel method for generating new kernels by solving differential equations for kernel functions. We examine specific kernels generated using this method. These kernels are applied to various data sets and compared against state-of-the-art kernels.