Title :
General memory neural network-extending the properties of basis networks to RAM-based architectures
Author :
Kolcz, Aleksander ; Allinson, Nigel M.
Author_Institution :
Dept. of Electron., York Univ., UK
Abstract :
A common framework for architectures combining multiple vector-quantization of the input space with memory look-up operations is proposed. Properties of the model are discussed and, in particular, a close relationship with basis functions networks (such as RBFs and kernel regression networks) is established
Keywords :
memory architecture; neural net architecture; random-access storage; vector quantisation; RAM-based architectures; RBFs; basis functions networks; basis networks; kernel regression networks; memory look-up operations; memory neural network; multiple vector-quantization; Computer architecture; Computer networks; Design methodology; Function approximation; Kernel; Neural networks; Random access memory; Robustness; Shape; Training data;
Conference_Titel :
Neural Networks, 1995. Proceedings., IEEE International Conference on
Conference_Location :
Perth, WA
Print_ISBN :
0-7803-2768-3
DOI :
10.1109/ICNN.1995.488864
