**What is Machine Learning?** You are probably thinking The Matrix or maybe The Terminator? Well, you are not too far off, as **Machine Learning** is a subset of artificial intelligence (AI) that enables computers to solve problems by using real-world data. This allows computers to continuously learn and improve without explicitly being programmed to do so.

While traditionally, a human element is required to solve a problem, with **Machine Learning** we move away from this approach. A problem solver (human element) is usually required to analyse and compute a solution for the problem before programming it. Now, with **Machine Learning**, the problem solver uses the general idea of what the problem is and how to solve it as a flexible component called a model, and uses an algorithm to adjust the model to real-world data. **Machine Learning** being an intersection of statistics, computer science, and applied mathematics automates the statistical reasoning and pattern-matching traditionally done by the problem solver. Remember, that the better the training data, the better the model will be.

**Machine Learning** has three learning subsets. What are they you ask?

**Supervised learning**: Every sample from the dataset has a corresponding label/output value associated with it. Thus, the algorithm learns to predict the labels/output values.
**Unsupervised learning**: The training data has no labels for the training data. The algorithm tries to learn the underlying patterns or distributions that govern the data.
**Reinforcement learning**: An algorithm figures out which actions to take in a situation to gain the maximum reward (a numerical value) to achieve the specific goal.

Seems like you've made it to the end of this boring blogπππ. Well, I hope you enjoyed it and you'll be following my machine learning journey. To keep up with my journey follow me on Hashnode and Twitter - I'll try to keep my posts short and to the point, also feel free to comment tips and notes for me. But for now, pick a pill (The Matrix reference - for all you young ones that might not know)!

]]>