Fraunhofer-Gesellschaft

Publica

Hier finden Sie wissenschaftliche Publikationen aus den Fraunhofer-Instituten.

Kernel-based generative learning in distortion feature space

 
: Tang, B.; Baggenstoss, P.M.; He, H.

:

Institute of Electrical and Electronics Engineers -IEEE-:
IEEE Symposium Series on Computational Intelligence, SSCI 2017
Piscataway, NJ: IEEE, 2018
ISBN: 978-1-5386-2726-6
ISBN: 978-1-5386-2727-3
ISBN: 978-1-5386-2725-9
pp.3341-3348
Symposium Series on Computational Intelligence (SSCI) <10, 2017, Honolulu/Hawaii>
English
Conference Paper
Fraunhofer FKIE ()

Abstract
This paper presents a novel kernel-based generative classifier which is defined in a distortion subspace using polynomial series expansion, named Kernel-Distortion (KD) classifier. An iterative kernel selection algorithm is developed to steadily improve classification performance by repeatedly removing and adding kernels. The experimental results on character recognition application not only show that the proposed generative classifier performs better than many existing classifiers, but also illustrate that it has different recognition capability compared to the state-of-the-art discriminative classifier - deep belief network. The recognition diversity indicates that a hybrid combination of the proposed generative classifier and the discriminative classifier could further improve the classification performance. Two hybrid combination methods, cascading and stacking, have been implemented to verify the diversity and the improvement of the proposed classifier. Experimental results show that our proposed generative Kernel-Distortion classifier has the best performance compared to the other four generative classifiers when combining with discriminative classifiers.

: http://publica.fraunhofer.de/documents/N-621890.html