DocumentCode :
3470987
Title :
Hardware design issues of fuzzy neural networks
Author :
Gobi, Adam F. ; Pedrycz, Witold
Author_Institution :
Dept. of Electr. & Comput. Eng., Alberta Univ., Edmonton, Alta., Canada
Volume :
2
fYear :
2004
fDate :
27-30 June 2004
Firstpage :
587
Abstract :
It has been well established that fuzzy neurons and fuzzy neural networks (FNN) are highly adaptive to changing conditions, possessing robust learning capabilities and inherent transparency for supporting a high level of knowledge interpretability. Consequently, they have the potential to provide exceptional mechanisms for building intelligent systems that must operate in dynamic and rapidly changing environments. However, to fully exploit the potential of FNN structures and their parallel nature, efficient hardware implementation techniques need to be developed. Here we are concerned with their realization using "standard" digital hardware so they may appeal to a wide range of applications, and our objective in this study is to investigate this avenue and identify various critical design issues.
Keywords :
fuzzy logic; fuzzy neural nets; neural net architecture; fuzzy neural networks; fuzzy neurons; hardware design; intelligent systems; knowledge interpretability; Artificial neural networks; Buildings; Fuzzy logic; Fuzzy neural networks; Fuzzy sets; Intelligent structures; Intelligent systems; Neural network hardware; Neurons; Robustness;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Fuzzy Information, 2004. Processing NAFIPS '04. IEEE Annual Meeting of the
Print_ISBN :
0-7803-8376-1
Type :
conf
DOI :
10.1109/NAFIPS.2004.1337367
Filename :
1337367
Link To Document :
بازگشت