Are artificial neural networks white boxes?

Author

Kolman, Eyal ; Margaliot, Michael

Author_Institution

Sch. of Electr. Eng.-Syst., Tel Aviv Univ., Israel

Volume

16

Issue

4

fYear

2005

fDate

7/1/2005 12:00:00 AM

Firstpage

844

Lastpage

852

Abstract

In this paper, we introduce a novel Mamdani-type fuzzy model, referred to as the all-permutations fuzzy rule base (APFRB), and show that it is mathematically equivalent to a standard feedforward neural network. We describe several applications of this equivalence between a neural network and our fuzzy rule base (FRB), including knowledge extraction from and knowledge insertion into neural networks.

Keywords

artificial intelligence; feedforward neural nets; fuzzy logic; knowledge acquisition; Mamdani type fuzzy model; all permutation fuzzy rule base; artificial neural networks; feedforward neural network; knowledge extraction; knowledge insertion; Artificial neural networks; Computer networks; Equations; Fuzzy neural networks; Fuzzy reasoning; Fuzzy sets; Fuzzy systems; Logistics; Mathematical model; Neural networks; Feedforward neural networks; hybrid intelligent systems; knowledge-based networks; rule extraction; rule generation; rule refinement; Algorithms; Computer Simulation; Computing Methodologies; Decision Support Techniques; Fuzzy Logic; Models, Biological; Models, Statistical; Neural Networks (Computer); Numerical Analysis, Computer-Assisted; Pattern Recognition, Automated; Stochastic Processes;

fLanguage

English

Journal_Title

Neural Networks, IEEE Transactions on

Publisher

ieee

ISSN

1045-9227

Type

jour

DOI

10.1109/TNN.2005.849843

Filename

1461427