The Artificial Umbrella

Source: Deep Learning on Medium

Right now, the tech industry is growing exponentially, and everyone is throwing around terms like Machine Learning, Deep Leaning, Artificial Intelligence, and what-the-heck algorithms.

But what do they mean? How are they related, if at all? And why are so, so many people talking about them?! I get it, you want to be in on this crazy AI secret, and so do I!

Artificial Intelligence

Let’s start with the basics and go over it together. What IS Artificial Intelligence, or AI?

In Simple Terms, AI is essentially the area of computer science that is attempting to make computer and machines smart. But what does that mean? Well, think of a computer having the same cognitive (or thinking), abilities as a human! Yes! A computer able to do the exact same things that you or I can do, and way more.

The thing about computers is that they’re actually way, way smarter than us. We go through life, experiencing ONE result of every decision we make. Don’t you ever wish we could experience all the potential results? Computers are able to do that because they can play out scenarios over and over and decide on the best plan of action.

You can refer to my other article on the basics of Artificial Intelligence to learn more about how deep AI really goes.

Doesn’t make sense yet? We’re going to cover the difference in this article.


When talking about AI, the word algorithm comes up very often. Algor-what? Isn’t that the word we used to solve a rubiks-cube ultra fast back in 6th grade? Well, almost.

An algorithm is basically just a pattern, or set of rules, that are followed during solving a problem, especially by a computer. A basic example being an algorithm for division.

But, what is an algorithm when it comes to AI? Since we now know that an algorithm tells a computer what to do and when, when connected together algorithms become stronger. An algorithm simply states the instructions through which a decision is made, where as AI uses trained data to make decisions. AI revolves around the use of algorithms.

Machine Learning

Now what does all this have to do with Machine Learning (ML)? Can machines even learn? The answer is yes. ML is an application of artificial intelligence. It allows systems to learn and improve based on experience, without being programmed to do something.

Artificial Intelligence is simply the broader umbrella that ML falls under, and helps machines to do things in a way we would consider “smart”.

There are huge applications for ML, such a medical diagnosis, and image processing.

Deep Learning

Now lets move on to Deep Learning… What is that? Deep Learning (DL) is really just a subset of machine learning, which is a subset of artificial intelligence. It’s all kind of nested, if you can’t tell. It is able to learn unsupervised from data that is unlabeled and unstructured.

It lets us train AI to predict outputs given a set of inputs. It uses multiple layers to progressively extract more information and patterns from any raw input. For example this could be used in image processing, where lower layer identify edges (basic), and higher layers identify more meaning data that can actually be useful to humans, such as numbers, letters, or faces.

What have we learned?

Throughout this article we’ve put together some basic information about what Artificial intelligence is, and what all these words really mean. Let’s take a look at our key takeaways

  • Artificial Intelligence is the process of making computers smart.
  • An Algorithm is the process used to solve a problem.
  • Machine learning is a subset under Artificial Intelligence that allows computers to learn for themselves
  • Deep Learning is a subset under Machine Learning that allows computers to learn more by themselves, from data that is unlabeled and unstructured, in layers.