Title :
Self-generating vs. self-organizing, what´s different?
Author :
Wen, W.X. ; Pang, V. ; Jennings, A.
Author_Institution :
Telecom Res. Lab., Clayton, Vic., Australia
Abstract :
Comparisons between the self-generating neural network (SGNN) and the self-organizing neural network (SONN) are performed. Although the SGNN concept is developed from SONN, the results obtained show that it has significant advantages when compared with SONN. These include simplicity of design methodology, greater speed for both training and testing, higher accuracy of clustering/classification, and better generalization capability. An analysis is conducted to investigate why SGNN is superior to SONN
Keywords :
generalisation (artificial intelligence); learning (artificial intelligence); neural nets; self-adjusting systems; classification; clustering; generalization; learning; self-generating neural network; self-organizing neural network; Algorithm design and analysis; Australia; Clustering algorithms; Design methodology; Humans; Neural networks; Neurons; Organizing; Telecommunications; Testing;
Conference_Titel :
Neural Networks, 1993., IEEE International Conference on
Conference_Location :
San Francisco, CA
Print_ISBN :
0-7803-0999-5
DOI :
10.1109/ICNN.1993.298773
