A high-level look at the biological neuron and what came next.
The fundamental goal of AI is for computer systems to achieve human-like intelligence. In the context of computer systems, we can define intelligence as the ability to perform complex tasks such as problem-solving, reasoning, perception, and communication.
To achieve this goal, researchers naturally looked to the human brain for inspiration. This led to the development of Artificial Neural Networks (ANN’s), which loosely model the biological neurons found in our brains.
I will outline the characteristics of biological neurons and highlight their influence over the early artificial neural networks developed in the mid-twentieth century. Don’t worry, I won’t over-complicate things. Instead, I’ll keep explanations high-level.
The nervous system, which is essentially the command center of the human body, is made up of neurons that send signals across the body. We use neurons to process and react to stimuli — let’s call this our inputs. Given some arbitrary input, the affected neurons must decide what action to take.
So what exactly is a neuron?
Neurons are messengers. They use electrical pulses and chemical signals to transmit information between different areas of the brain and between the brain and the rest of the nervous system via connections called synapses.
There are three components of the neuron, shown in the image above, that are useful to us. The dendrites, which receive signals, can be thought of as the inputs to the neuron. The cell body — or soma — is the central hub of the neuron. Finally, the axon, which we can think of as the output.
If the number of inputs exceeds the activation threshold, the neuron is triggered. When triggered, it sends an impulse out of the neuron through the axon.
That’s a biological neuron in a nut-shell. As I said, it’s a very high-level description of biological neurons. If you’re after more details, I suggest looking for a more in-depth guide.