Just like the title implies, we are not going deep into the complexity of machine learning. This is a brief intro of what machine learning is, it's main types, and applications.

Let's get started!🤩

What is machine learning?

Machine learning is simply the art of extracting knowledge from data. With machine learning, computers are able to learn from data and perform some boring or complex tasks without following explicit instructions. This is interesting, right?😀

Now, the question is: where is it actually applied? 🤔
Machine learning has made a significant impact in our everyday life even if sometimes we don't know this. From automatic recommendation of which videos to watch, which products to buy, to recognising your friends in photos.

Many modern websites and devices have machine learning at their core. When you look at a complex website like Facebook, Youtube, Amazon, or Netflix, it is very likely that every part of the site contains multiple machine learning models(Don't worry about some terms☺️) A machine learning model is simply a file/program that has been trained to recognise a pattern from data.

Let's look at different problems machine learning can solve😉

Real world problems are not solved by a single fixed algorithm, every problem has its own way to solve it depending on the data available and expected output, so it's up to you as a machine learning engineer to choose the right algorithm to use. They are very many machine learning algorithms and these algorithms are also classified into 3 main types.

1. Supervised Learning
Supervised learning is one of the most commonly used and successful type of machine learning. Supervised learning is used when we already have examples of input and their output and we need to build a machine learning model that will predict the output of a new given input.

An example of supervised machine learning task:

• Detecting fraudulent activity in credit card transaction: Here the input is a record of the credit card transaction, and the output is whether this transaction is likely to be fraudulent or not.

2. Unsupervised Learning
Unsupervised learning is simply used when we only have input data and no output data given to the algorithm.

An example of unsupervised machine learning task:

• Identifying topics is a set of blog posts: Let's say you have a large collection of text data, you might want to summarise it and find relevant themes in it. You might not even know what these topics are or how many topics there might be and therefore, they are no known outputs.

3. Reinforcement Learning
🤩This one is another level of machine learning👌. The system learns just like a human does(Learn by doing😀). The learning system called an agent can observe the environment, select and perform actions, and get rewards in return when it made a good choice or penalties when it did not do well. It might then learn by itself what is the best strategy called policy to get the most rewards over time.

An example of Reinforcement machine learning:

• DeepMind's AlphaGo program: which made the headlines in March 2016 when it beat the world champion Lee Sedol at the game of Go. It learned its winning policy by analysing millions of games, then playing many games against itself. Finally, it won 💪🤩 by just applying the policy it had learned.

This is a little longer than I expected 🥱 but I couldn't bother providing the full-package I had for you.👇

🟢 An advice: When going deep into the aspects of of machine learning, it is easy to lose sight of your goals. This field is very huge and they are so many ways to solve a particular problem, it is good to keep a big picture in mind all the time.

Good luck 🥳
See ya on my next article about "How you can build your first machine learning model"😀
Feel free to comment down below👇 about anything.