Title :
Inverse-step competitive learning
Author :
Yin, Hao ; Lengelle, R. ; Gaillard, Paul
Author_Institution :
Genie Inf., Compiegne Univ., France
Abstract :
Reviews several variants of the competitive learning rule: simple competitive learning, the Kohonen self-organization map, and frequency-sensitive competitive learning. They then propose a novel learning rule based on competitive learning, called inverse-step competitive learning (ISCL). The isolated points play a more important role than the normal points in simple competitive learning, because the learning step is proportional to the distance between the input pattern and the weights value. The basic idea of this learning rule is to take a learning step which is a descending function of this distance. The authors give the first results of the ISCL rule and compare it to simple competitive learning
Keywords :
learning systems; neural nets; pattern recognition; Kohonen self-organization map; competitive learning rule; descending function; frequency-sensitive competitive learning; input pattern; inverse-step competitive learning; weights value; Algorithm design and analysis; Backpropagation algorithms; Clustering algorithms; Computer architecture; Frequency; History; Neural networks; Pattern clustering; Pattern recognition;
Conference_Titel :
Neural Networks, 1991. 1991 IEEE International Joint Conference on
Print_ISBN :
0-7803-0227-3
DOI :
10.1109/IJCNN.1991.170505