Title :
Bipolar pattern association using a recurrent winner take all network
Author :
McInroy, John E. ; Wilamowski, Bogdan M.
Author_Institution :
Dept. of Electr. Eng., Wyoming Univ., Laramie, WY, USA
Abstract :
A neural network for heteroassociative (or autoassociative) pattern recognition of input bipolar binary vectors is proposed. By combining the advantages of feedforward and recurrent techniques for heteroassociation, a simple network with guaranteed error correction is found. The heart of the network is based on a new, recurrent method of performing the winner-take-all function. The analysis of this network leads to design rules which guarantee its performance. The network is tested on a character recognition problem utilizing the entire IBM CGA character set
Keywords :
content-addressable storage; feedforward neural nets; pattern recognition; recurrent neural nets; IBM CGA character set; autoassociative pattern recognition; bipolar pattern association; character recognition; feedforward techniques; guaranteed error correction; heteroassociative pattern recognition; input bipolar binary vectors; recurrent techniques; recurrent winner-take-all network; Character recognition; Error correction; Hamming distance; Heart; Network topology; Neural networks; Pattern recognition; Performance analysis; Recurrent neural networks; Testing;
Conference_Titel :
Neural Networks,1997., International Conference on
Conference_Location :
Houston, TX
Print_ISBN :
0-7803-4122-8
DOI :
10.1109/ICNN.1997.616209
