EVALUATION OF THE SIMPLEX-METHOD FOR TRAINING SIMPLE MULTILAYER NEURAL NETWORKS

A learning algorithm based on the modified Simplex method is proposed for training multilayer neural networks. This algorithm is tested for neural modelling of experimental results obtained during crossflow fil tration tests. The Simplex method is compared to standard back-propaga tion. Simpler to implement, Simplex has allowed us to achieve better r esults over four different databases with lower calculation times. The Simplex algorithm is therefore of interest compared to the classical learning techniques for simple neural structures.