Different approaches for Machine Learning

Different approaches for Machine Learning in an informal way.

You may think of machine learning as a different name for something oh no it's already around us. Maybe it's just an updated way to describe statistics or novel ways to discuss data science. Machine learning certainly has statistics! It might be also a key part of your data effort. But machines need to learn as humans learn new things. Think about what it means to learn? What are the different methods that you use to learn?

Suppose you wanted to learn how to play chess. You may do this in a couple of different ways. You may hire a chess tutor or let's suppose you couldn't find a tutor then you may go to public parks and watch several hundred experts play the game obviously you are not allowed to ask questions you have just quietly watched and learned. If you do this long enough you have probably understood the game. You might not know the name of chess pieces, but you could understand the moves and strategies from your observations. Or you might even try the combination of these two approaches.

These three strategies are very similar to how machines might learn. You could do something called supervised learning, here data scientists act as a tutor for the machines. They train the machine by showing the basic rules giving it an overall strategy. You could also do unsupervised learning here you just let the machine make all the observations on its own. The machine might not know all the different names and labels however they will find patterns on their own. Finally, you could mix the two methods and have semi-supervised learning. Here you will need to train the machine just a little bit so they get a high-level overview and the most part of learning about the rules and strategies is through observing different patterns. As you can imagine all three approaches has their pros and cons.