Neural Networks and Model Inversion

An artificial neural network consists of two general elements the processing units, called neurons, and the connections between the neurons, the weights (Fig. 3.6). The weights are the analogs of the synapses in biological networks and are the main 

It has asymptotic limits of 0 and 1 for very large negative and very large positive Sj. 

The output of the neuron (a number in the range between 0 and 1) is transmitted to a following neuron multiplied by their joint weight. 

---