We derive a novel derivative based version of kernelized Generalized Learning Vector Quantization (KGLVQ) as an effective, easy to interpret, prototype based and kernelized classifier. It is called D-KGLVQ and we provide generalization error bounds, experimental results on real world data, showing that D-KGLVQ is competitive with KGLVQ and the SVM on UCI data and additionally show that automatic parameter adaptation for the used kernels simplifies the learning.
|Title of host publication||Proc. Intelligent Data Engineering and Automated Learning - IDEAL 2010|
|Editors||Colin Fyfe, Peter Tino, Darryl Charles, Cesar Garcia-Osoro, Hujun Yin|
|Number of pages||8|
|Publication status||Published - 2010|
|Name||Lecture Notes in Computer Science|