Title :
A parallel approach to plastic neural gas
Author :
Ancona, Fabio ; Rovetta, Stefano ; Zunino, Rodolfo
Author_Institution :
Dept. of Biophys. & Electron. Eng., Genoa Univ., Italy
Abstract :
A parallel implementation of unsupervised vector-quantization networks can reduce the high computational load of the training process. First, a plastic version of the neural gas algorithm is presented. Then, the paper describes how a toroidal mesh topology fits the neural model for a distributed implementation. The architecture adopted and the data-allocation strategy enhance the method´s scaling properties and remarkable efficiency. Experimental results on a significant testbed (low bit-rate image compression) confirm the validity of the parallel approach
Keywords :
neural nets; parallel algorithms; unsupervised learning; vector quantisation; data-allocation strategy; low bit-rate image compression; parallel implementation; plastic neural gas; toroidal mesh topology; unsupervised vector-quantization networks; Clustering algorithms; Concurrent computing; Electronic mail; Image coding; Iterative algorithms; Neurons; Plastics; Prototypes; Testing; Vector quantization;
Conference_Titel :
Neural Networks, 1996., IEEE International Conference on
Conference_Location :
Washington, DC
Print_ISBN :
0-7803-3210-5
DOI :
10.1109/ICNN.1996.548878
