ABSTRACT
This chapter provides the optimal number of hidden-layer neurons of the constructed two-input Legendre-polynomial neuronet, two weights-and-structure-determination (WASD) algorithms of growing type are designed on the basis of the weights-direct-determination (WDD) subalgorithm. In addition, based on theWDD subalgorithm, two WASD algorithms with different growing speeds are built up to determine the optimal weights and structure of the two-input Legendre-polynomial neuronet. Numerical results further substantiate the efficacy of the two-input Legendre-polynomial neuronet equipped with either of such two WASD algorithms. Numerical studies further verify the efficacy of the two-input Legendre-polynomial neuronet equipped with the two aforementioned WASD algorithms. It explores numerical studies are performed to verify the efficacy of the two-input Legendre-polynomial neuronet equipped with one of the aforementioned two WASD algorithms. Based on the WDD subalgorithm, two WASD algorithms are presented to further improve the performance of the two-input Legendre-polynomial neuronet.
