Numerous methods based on distance computation are available to carry out vectors comparison. Unfortunately, most of them are applicable only if the vectors are of the same length or do not take into account components' misalignment. This paper presents a new distance between two representations called the Elastic Distance and based on the dynamic programming technique. Properties are studied. We show that it leads to a variant of the Least Vector Quantisation technique that learns the best representants of a group of prototypes. A new centroid computation algorithm is proposed. Finally, the learning scheme algorithm has been successfully applied on an on-line numerical handwritten character recognition problem using a previously computed centroid of a set of prototypes.
[1]
Y. L. Cun,et al.
Modèles connexionnistes de l'apprentissage
,
1987
.
[2]
Bernard Widrow,et al.
Adaptive switching circuits
,
1988
.
[3]
Teuvo Kohonen,et al.
Self-Organization and Associative Memory
,
1988
.
[4]
S. Chiba,et al.
Dynamic programming algorithm optimization for spoken word recognition
,
1978
.
[5]
Shun-ichi Amari,et al.
A Theory of Adaptive Pattern Classifiers
,
1967,
IEEE Trans. Electron. Comput..
[6]
Michael J. Fischer,et al.
The String-to-String Correction Problem
,
1974,
JACM.
[7]
Teuvo Kohonen,et al.
Self-organization and associative memory: 3rd edition
,
1989
.