Hier finden Sie wissenschaftliche Publikationen aus den Fraunhofer-Instituten.

Joint Selection of Central and Extremal Prototypes Based on Kernel Minimum Enclosing Balls

: Bauckhage, Christian; Sifa, Rafet


Singh, L. ; Institute of Electrical and Electronics Engineers -IEEE-; IEEE Computer Society:
IEEE International Conference on Data Science and Advanced Analytics, DSAA 2019. Proceedings : Washington, District of Columbia, United States, 5-8 October 2019
Los Alamitos/Calif.: IEEE Computer Society Conference Publishing Services, 2019
ISBN: 978-1-7281-4493-1
ISBN: 978-1-7281-4494-8
International Conference on Data Science and Advanced Analytics (DSAA) <2019, Washington/DC>
Conference Paper
Fraunhofer IAIS ()
Kernel Minimum Enclosing Balls; Prototype Extraction; unsupervised learning; Frank Wolfe Algorithm

We present a simple, two step procedure that selects central and extremal prototypes from a given set of data. The key idea is to identify minima of the function that characterizes the interior of a kernel minimum enclosing ball of the data. We discuss how to efficiently compute kernel minimim enclosing balls using the Frank-Wolfe algorithm and show that, for Gaussian kernels, the sought after prototypes can be naturally found via a variant of the mean shift procedure. Practical results demonstrate that prototypes found this way are descriptive, meaningful, and interpretable.