Comparison between MLP and RBF network in improving CHEMFET sensor selectivity

Unit composite friction coefficient of model pile floated in kaolin clay reinforced by recycled crushed glass under uplift loading

“…(2) Note: Goal: MSE, spread: spread of the Gaussian function, MAXN: maximum number of neurons in hidden layer, DisplayAt: neuron numbers to add between displays. c i : center of the ith hidden nodes, w ij : connection weights, 𝜑 𝑖 : RBF function (Gaussian function) [31] [32] [33] [34] [35]. The topology of RBF is the same as single layer perceptron, but the distance from the center.…”