Many of us keep hearing (and using) the term Machine Learning yet, very few of us truly understand the meaning of it.

Don’t worry, it’s not rocket science. Here are the basics of Machine Learning

Simplifying Machine Learning

Since we are talking about basics, let us start with a real-life example.

Imagine you wake up every morning at 7 am sharp. Next, you switch on the geyser for a hot shower and then dress up in the next 15 minutes. Then, you drink a cup of steaming espresso around 7:30 am and read the newspaper for the next 15 minutes. After which you book a cab and leave.

Now here’s another scenario. Imagine there is already warm water running in the shower when you wake up. Your cup of espresso is already served by the time you are ready. You have a machine that reads the latest headlines to you and books a cab for you with live traffic updates.

Yes, this automation of your life is something that’s made by possible by machine learning, which is nothing but a set of algorithms that are fed into a computer that makes data-driven decisions and evolves when exposed to new data over a period of time.

It is a part of Artificial Intelligence and Data Science, which studies certain patterns, behaviours, and figures out the results based on the continuous data that it gathers.

What are the different kinds of Machine Learning?

There are 3 main broad categories

Supervised Learning: This is classic textbook style learning where a teacher explains the difference between apples and oranges with certain labels or pictures. Similarly, when we input certain labels into a computer, we get the result with the desired output.

Unsupervised Learning: This is more of a DIY (Do It Yourself) learning approach where algorithms study various combinations and permutations of the data that’s fed into the computer. The output is based on the clusters or groups of information without any exclusive labels.

Reinforcement Learning: Let’s take chess as an example. Certain moves, allow you maximum gains and improve your chances of winning. Imagine if AI can track all your moves,  then based on certain patterns, feedback and outcomes it will be able to derive a model, which will reinforce you to apply those moves again.

So there you have it. A quick guide to understanding machine learning. Stay tuned for more. Or check out Google’s Machine Learning Crash Course