An illustration of a monkey’s neurons activating when it sees an action. The action in question is another monkey eating a banana.

The human neural network #

The human brain has around 86 billion neurons.

Neurons are basic building blocks used to exchange info with other neurons via electrical pulses.

One neuron doesn’t talk directly with all 80 billion others. Rather, they are connected into structures that perform specialized functions. These structures are biological neural networks.

Brains learn from experience. When you learn something new, neurons in your brain strengthen their lines of connection with some neurons and prune their connection to others.

Artificial neural networks #

Humans perceive and interpret everything based on their previous experiences. We make decisions using reasoning and consider all of our life experiences. On the other hand, computers lack the ability to understand the meaning of data in the same way humans do. They can only follow instructions.

However, computers can learn through training. Artificial neural networks learn through trial and error (AKA failing a lot). These errors are then used to adjust the connections between neurons so that the network can be more correct in the future. See how this process is similar to how humans learn?

Continue this process hundreds or thousands of times, and you’ll end up with a somewhat smart neural network, capable of making complex decisions.

Neurons are like functions #

Neurons perform math/programming functions and take parameters (x1, x2, …, xn) to output a result. Having more neurons makes it possible to perform more complex tasks.

TensorFlow Playground is a cool visualization of neurons working together to solve a classification problem.