Cognilytica is now part of PMI!

When referring to neural networks, a node (or neuron) has one or more weighted input connections, a transfer function combining the inputs in some way, and an output connection. The sum of the products of any input weights is calculated in each node, and if the value is above some threshold, the neuron fires and takes the activated value, otherwise it takes the deactivated value. The activation threshold is usually above 0, the activated value is usually 1, and the deactivated value is usually -1. An activation function calculates the activation threshold.