On Some Properties of a Hetereogeneous Transfer Function Involving Symmetric Saturated Linear (SATLINS) with Hyperbolic Tangent Sigmoid (TANSIG) Transfer Functions
Christopher Godwin Udomboso · Progress in applied mathematics · 2013
Transfer functions maps the input layer of the statistical neural network model to the output layer. To do this perfectly, the function must lie within certain bounds. This is a property of probability distributions. This paper establishes the heterogeneous transfer function, SATLINS_TANSIG, as a Probability Distribution Functions (p.d.f) by showing that it is proper. It also shows the mean and variance.