Title :
Bit Importance Strategy For Bidirectional Associative Memory
Author :
Wei, Mo ; Cruz, Jose B., Jr.
Author_Institution :
Life Fellow, IEEE, Ohio State University, Columbus, OH 43210 USA (email: wei.73@osu.edu).
Abstract :
In this paper, a strategy named Bit Importance Strategy (BIS) is introduced to improve the performance of Bidirectional Associative Memory (BAM) in cases in which differences in importance among bits must be considered. The main advantage of this new strategy lies in the improvement of the recall success ratio. Simulation results are given, with an application to a simplified version of a military battlefield operation, based on a theoretical analysis.
Keywords :
Analytical models; Artificial neural networks; Associative memory; Delay; Encoding; Filters; Magnesium compounds; Noise level; Nonlinear equations; Signal to noise ratio;
Conference_Titel :
Decision and Control, 2005 and 2005 European Control Conference. CDC-ECC '05. 44th IEEE Conference on
Print_ISBN :
0-7803-9567-0
DOI :
10.1109/CDC.2005.1582143