|
Published Articles >> Table of Contents >> Abstract
IEEE-INNS-ENNS International Joint Conference on Neural Networks (IJCNN'00)-Volume 1
p. 1106
The K-Winner Machine Model
Sandro Ridella, University of Genoa
Stefano Rovetta, University of Genoa
Rodolfo Zunino, University of Genoa
Full Article Text:
  
DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/IJCNN.2000.857822
Send link to a friend
| Abstract |
|
A K-Winner Machine (KWM) selects among a family of classifiers the specific configuration that minimizes the expected generalization error. In training, KWM uses unsupervised Vector Quantization and subsequent calibration to label data-space partitions. At run time, KWM seeks the largest set of best-matching prototypes agreeing on a test sample, and provides a local-level measure of confidence. The VC-dim of a KWM classifier is worked out exactly; the resulting small values set tight bounds to generalization performance. The method applies to high-dimensional, multi-class problems with large data sets. Experimental results on both a synthetic and a real domain (NIST handwritten numerals) validate confirm the consistency of the theoretical framework.
|
 Additional Information
|
Citation:
Sandro Ridella, Stefano Rovetta, Rodolfo Zunino,
"The K-Winner Machine Model,"
ijcnn,
p. 1106,
IEEE-INNS-ENNS International Joint Conference on Neural Networks (IJCNN'00)-Volume 1,
2000
|
|
