ARE ARTIFICIAL NEURAL NETWORKS BLACK-BOXES

Artificial neural networks are efficient computing models which have s hown their strengths in solving hard problems in artificial intelligen ce. They have also been shown to be universal approximators. Notwithst anding, one of the major criticisms is their being black boxes, since no satisfactory explanation of their behavior has been offered. In thi s paper, we provide such an interpretation of neural networks so that they will no longer be seen as black boxes. This Is stated after estab lishing the equality between a certain class of neural nets and fuzzy rule-based systems. This interpretation is built with fuzzy rules usin g a new fuzzy logic operator which is defined after introducing the co ncept of f-duality, In addition, this interpretation offers an automat ed knowledge acquisition procedure.