Title :
Generative vector quantisation
Author :
Westerdijk, Machiel ; Barber, David ; Wiegerinck, Wim
Author_Institution :
Dept. of Med. Phys. & Biophys., Nijmegen Univ., Netherlands
Abstract :
Based on the assumption that a pattern is constructed out of features which are either fully present or absent, we propose a vector quantisation method which constructs patterns using binary combinations of features. For this model there exists an efficient EM-like learning algorithm which learns a set of representative codebook vectors. In terms of a generative model, the collection of allowed binary states `generates´ the set of codebook vectors. Thus, the method provides not only a compact description of the data in terms of clusters, but also an explanation of the individual clusters in terms of common elementary features. Preliminary results on image compression and handwritten digit analysis indicate that our approach is a computationally inexpensive alternative to more complex probabilistic generative graphical models
Keywords :
learning (artificial intelligence); binary states; clusters; codebook vectors; feature extraction; generative vector quantisation; handwritten digit analysis; image compression; learning algorithm;
Conference_Titel :
Artificial Neural Networks, 1999. ICANN 99. Ninth International Conference on (Conf. Publ. No. 470)
Conference_Location :
Edinburgh
Print_ISBN :
0-85296-721-7
DOI :
10.1049/cp:19991232