Artificial Vs Biological Neural Networks

A Biological Neural Network is comprised of a giant network, or graph, of neurons comprised of a Soma, a nucleus, dendrites that accept the input signal, and axom hillock that in an oversimplified way determines the trigger and an axon that usually branches to carry the output signal. Artificial neural networks abstract these biological entities with an input signal with weights that mimic the dendrite, an activation function that mimics the axon hillock and simple messaging that mimics the axon.

Biological Neural Networks Frequency Encode Their Signals

Since the intensity of the signal experienced by the axon hillock (triggerzone) depends on how many spikes at that moment excite that region at the same time, the signal is based not only on how many spikes there are, but also on the shape of the neuron and the timing of the signals. The neuron performs a spatio- temporal integration of the incoming signals. If the excitation level at a given time surpasses its threshold, an action potential is generated and passed down the axon.

Artificial Neural Networks Amplitude Encode Their Signals

With artificial NNs, the neurons amplitude encode their signals. Temporal considerations are less important here. The models mimic and encapsulate the strength and power of the entire signal, not the quantity of small signals over time.

Does It Matter?

Wet biological systems are the way they are because they were the first system that worked from random evolutionary mutations. So, just like we don't need to flap the wings on a jet airplane to achieve artificial controlled flight, maybe we really don't need to worry about frequency or amplitude encoding.