Note: Your custom schedule will not be saved unless you create a new account or login to an existing account.
| Paper: | MLSP-P7.9 |
| Session: | Pattern Recognition and Classification II |
| Time: | Friday, May 21, 15:30 - 17:30 |
| Presentation: |
Poster |
| Topic: |
Machine Learning for Signal Processing: Signal detection, Pattern Recognition and Classification |
| Title: |
PROTOTYPE-BASED MINIMUM ERROR CLASSIFIER FOR HANDWRITTEN DIGITS RECOGNITION |
| Authors: |
Roongroj Nopsuwanchai; University of Cambridge | | |
| | Alain Biem; IBM T. J. Watson Research Center | | |
| Abstract: |
This paper describes an application of the prototype-based minimum classification error classifier (PBMEC) to the offline recognition of handwritten digits. The PBMEC uses a set of prototypes to represent each digit along with an Lp-norm of distances as the decoding scheme. Optimization of the system is based on the Minimum Classification Error criterion (MCE). In this paper, we introduce a new clustering criterion adapted to the PBMEC structure that minimizes an Lp norm-based distortion measure. The new clustering algorithm can generate a smaller number of prototypes than the standard k-means with no loss in accuracy. It is also shown that the PBMEC trained with the MCE realizes more than a 42% improvement from the baseline k-means process and requires only 28Kb storage to match the performance of a 1.46MB-sized k-NN classifier. |
| |
| Back | |