Hier finden Sie wissenschaftliche Publikationen aus den Fraunhofer-Instituten.

Towards an empirical and theoretical evaluation of gradient based approaches for finding kernel minimum enclosing balls

: Kondratiuk, H.; Sifa, R.


Webb, G. ; Institute of Electrical and Electronics Engineers -IEEE-; IEEE Computer Society:
IEEE 7th International Conference on Data Science and Advanced Analytics, DSAA 2020. Proceedings : 6-9 October 2020, Sydney, Australia
Los Alamitos, Calif.: IEEE Computer Society Conference Publishing Services (CPS), 2020
ISBN: 978-1-7281-8206-3
ISBN: 978-1-7281-8207-0
International Conference on Data Science and Advanced Analytics (DSAA) <7, 2020, Online>
Fraunhofer IAIS ()

In this paper we introduce a projected gradient descent algorithm to find kernel minimum enclosing balls and compare it to a gradient based Frank-Wolfe algorithm. We base our comparison on empirical as well as theoretical observations of the two methods by comparing different aspects of their behaviors that involve runtime, stability, abilities to find novel datapoints as well as convergence rates.