Neurophysiology and Artificial Neural Networks
I’m not a doctor in medicine, but I’m really fascinated by biology and especially how animals work. The only real pieces of intelligence we know so far are animals. That’s where almost all our inspiration in AI comes from.
Animals, for a large part, have a sensory and/or nervous system, and sometimes a central system to analyze information deeply. They are agents evolving in an environment, which only “goal” is to survive — it is implicit since this survival behaviour is embedded in the world, you can’t escape it (otherwise, you’re dead). Evolution has progressively selected the most “fitted” behaviours, to reach the species we have today.
It appears that, for most species, this nervous system can be described as a set of neurons, of different forms, communicating with each other. Our artificial neurons are completely inspired by this, and we’ll see how in more detail.
The biological neuron
We can describe a neuron as a compute unit, handling information and sending it to neighbours. In the real world around us, a neuron is quite complicated and doesn’t have a unique structure. But we will focus on the main characteristics.
A neuron has many receptors, the dendrites, and a unique canal to send the information : the axon. The nucleus is also called…
