Breaking down Machine Learning

I feel like machine learning is so popular in the tech field that even people outside the field have heard of it. At the very least, I would be quite surprised if most people have not heard AI (or Artificial Intelligence) at least mentioned in the past few years. For those who have not heard of machine learning, it’s a type of artificial intelligence with a not-so-agreed-upon definiton. Many people and entities argue about a precise definition, but I really liked the definition found here,
Machine learning is an application of artificial intelligence (AI) that provides systems the ability to automatically learn and improve from experience without being explicitly programmed.
Anyone new to machine learning is probably thinking — wait, what… machines… learning? How is that even possible? If you are thinking that right about now, you are completely justified in thinking that. I know I was dumbfounded when I first heard about machine learning. Heck, I am still taken aback by the fact that machine learning is a thing.
Even though machine learning is presented in this uber complex way (don’t get me wrong, the details get very complex), the idea is quite simple. Chances are that anyone that has taken a math class has heard of a function,

A function is simply a special relationship where each input (or set of inputs) has a single output. A function maps the input to the output. In machine learning, the computer is getting taught how to map a certain input (or set of inputs) to a corresponding output. Usually, there is no single input that is fully indicative of how to map to an output, so the function representation used in Figure 1 would look something like this,

Again, more granular details of machine learning get quite complex quite quickly, but I hope this was a helpful conceptual overview of what machine learning is.
Cheers!
