An effective learning of neural network by using RFBP learning algorithm

The Transmittance Estimation of Two Layers Coating TP Film by Neural Network

“…The error back-propagation (BP) learning algorithm is adopted for network's training. The major steps of BP algorithm could be summarized as follows [5,8,11,12].…”

“…In last two decades, neural network (NN) has been widely applied into the various science and management areas due to its powerful learning and mapping capabilities [5,6,7,8,9,10,11,12]. The complex relationship between input and output pairs of the training data could be easily developed through a simple training process.…”

A modified gradient-based neuro-fuzzy learning algorithm and its convergence

“…Neuro-fuzzy models have been constructed by combining fuzzy systems and neural networks to generate or tune fuzzy rules of fuzzy system models [2,[9][10][11]14,20,22,29,40], and recently have been successfully applied to, e.g. control system and system identification [5,7,16,18,21,23,24,31,36]. There are varieties of system structures and learning algorithms available for neuro-fuzzy models [11,20,37,3,13,15,17].…”

Modified constrained learning algorithms incorporating additional functional constraints into neural networks

“…However, these learning algorithms have not considered the network structure and the involved problem properties, thus their capabilities are limited [1,2,5,6,18]. In order to obtain better generalization capability [8,24], many constrained learning algorithms (CLA) incorporating additional functional constraints into neural networks have been proposed in literatures [4,[12][13][14][15]19,23,27].…”