Training fuzzy number neural networks using constrained backpropagation
James Dunyak, Murat Guven, Donald C. Wunsch · 2002
Few training techniques are available for neural networks with fuzzy number weights, inputs, and outputs. Typically, fuzzy number neural networks are difficult to train because of the many /spl alpha/-cut constraints implied by the fuzzy weights. In this paper, we introduce a weight representation that simplifies the constraint equations. A constrained form of backpropagation is then developed for fuzzy number neural networks. Standard backpropagation may be viewed as a constrained optimization of the linearization of the weight function. Our weight representation allows use of the additional /spl alpha/-cut constraints during a weight update.