Machine Learning (ML) is a technique that uses algorithms to learn from the data without being programmed explicitly. Due to the data abundance and efficient data storage, ML rose to the limelight in recent times, but the foundational research in this field was done in seventy’s and eighty’s.

Different ways for a computer to learn from data — supervised learning, unsupervised learning, and reinforcement learning.

A supervised learning algorithm takes labeled data while training the model, and then the model makes predictions in the presence of the new data. These problems could be divided into regression and classification problems.

**Classification**: When the target variable is categorical, such as identifying spam and non-spam mail.**Regression**: When the target variable is a real value, such as house price.

Unsupervised learning is when we are dealing with training data that is unlabelled that is without a target variable. The goal is to find patterns in the data to derive insights from it.

Two forms of unsupervised learning problems are **Clustering** and **Association.**

**Clustering**: A clustering problem is where you want to discover the groupings in the data based on a predefined similarity or distance metric in the feature space. For Example, finding customer segments in data.**Dimensionality Reduction:**It is a problem that uses an unsupervised learning technique where the number of random variables (under consideration)is reduced by projecting the feature space to a lower-dimensional space.**Association**: An association rule learning problem is where you want to find rules that majorly describes the data, such as people that buy Z also tend to buy V.

Reinforcement Learning(RL) is a type of machine learning technique that enables the agent to learn by trial and error using **rewards and punishments** as signals for **positive** and **negative behavior**. The goal is to find a suitable action model that would maximize the **total cumulative reward** of the agent.

Generally, reinforcement learning algorithms begin with a more explorative approach and as the reward systems are better understood, the algorithm will then lean towards exploitation.In the reinforcement problems, the act of reevaluating the probability in each state is known as a **Markov Decision Process (MDP)**.An MDP consists of a set of finite environment states S ,a set of possible actions A(s) in each state, a real valued reward function R(s) and a transition model P(s’, s | a).

**AlphaGoZero** computer program uses** RL** to defeat a world champion in the ancient Chinese game of Go.

Click 💚if you like the article.If you have any questions, you can write in the comments section below, and I will do my best to answer them.