There are 3 types of ML.

1. Supervised Learning – Train Me!
2. Unsupervised Learning – I am self sufficient in learning or Identify Clusters or Identify group
3. Reinforcement Learning – My life My rules! or Learn from mistakes or Hit & Trial (that is to guess a solution and see if it is valid or not)

1. Supervised Learning:- Supervised learning is when the model is getting trained on a labelled dataset. Labelled dataset is one which have both input and output parameters. For example: 12+8 =20. Here our input data (12 and 8) and output data is 20. 

It is the one, where you can consider the learning is guided by a teacher. We have a dataset which acts as a teacher and its role is to train the model or the machine. Once the model gets trained it can start making a prediction or decision when new data is given to it.

Supervised learning algorithms try to model relationships and dependencies between the target prediction output and the input features such that we can predict the output values for new data based on those relationships which it learned from the previous data sets.

There are many applications of Supervised Machine Learning. Some of its famous applications are:

Face Detection: A Supervised Machine Learning algorithm is used in face detection in which the machine is trained to detect your face.

Cortana: Your personal assistant “Cortana” is trained to detect human voice.

Types of Supervised Learning

i). Regression (no labels define)

The goal is to predict continuous values, e.g. home prices, Weather forecasting, Market Trends,etc.

Here are some popular Regression algorithms which come under supervised learning:

• Linear Regression
• Regression Trees
• Non-Linear Regression
• Bayesian Linear Regression
• Polynomial Regression

ii). Classification (defined labels)

The goal is to predict discrete values, e.g. {1,0}, {True, False}, {spam, not spam}. It is a Supervised Learning task where output is having defined labels(discrete value).

For example in Figure Classification, Output – Purchased has defined labels i.e. 0 or 1 ; 1 means the customer will purchase and 0 means that customer won’t purchase. The goal here is to predict discrete values belonging to a particular class and evaluate on the basis of accuracy.

It can be either binary or multi class classification. In binary classification, model predicts either 0 or 1 ; yes or no but in case of multi class classification, model predicts more than one class.

Examples: Risk Assessment, Image classification, Fraud Detection, spam filtering, Gmail classifies mails in more than one classes like social, promotions, updates, forum. etc.

Here are some popular Classification algorithms which come under supervised learning:

• Random Forest
• Decision Trees
• Logistic Regression
• Support vector Machines

2.Unsupervised Learning:- It is a type of Machine Learning that converts the unorganized data to organized data. The computer is trained with unlabeled data. It is mainly used in pattern detection and descriptive modeling.

Here there’s no teacher at all, actually the computer might be able to teach you new things after it learns patterns in data, these algorithms a particularly useful in cases where the human expert doesn’t know what to look for in the data.

The model learns through observation and finds structures in the data. Once the model is given a dataset, it automatically finds patterns and relationships in the dataset by creating clusters in it. What it cannot do is add labels to the cluster, like it cannot say this a group of apples or mangoes, but it will separate all the apples from mangoes.

Suppose we presented images of different types of fruits. Now the Unsupervised Learning algorithm will group the same kind of fruits and form clusters. Moreover, it separates one type of fruit from another type and gives an easy to understand interface to the user. So, Unsupervised Learning converts unorganized data to the organized form.

There are many examples of Unsupervised Machine Learning. Some of its famous examples are:

YouTube Video Recommendation System: Using your watch history YouTube traces the type of videos you are liking. After that, It also traces the other users that watched similar videos as you did and enjoyed other videos as well. Now, the YouTube recommendation system sees the relationship in the data. And finally, it gives video suggestions.

Facebook Friends or group or Page Recommendation System.

Ecommerce Website inter related Products Recommendation System.

It can be further classifieds into two categories of algorithms:

i)  Clustering: Clustering is a method of grouping the objects into clusters such that objects with most similarities remains into a group and has less or no similarities with the objects of another group.

ii) Association: An association rule is an unsupervised learning method which is used for finding the relationships between variables in the large database. It determines the set of items that occurs together in the dataset. Association rule makes marketing strategy more effective. Such as people who buy X item (suppose a laptop) are also tend to purchase Y (Laptop bag/Mouse/Keyboard) item.

some popular unsupervised learning algorithms:

• K-means clustering
• KNN (k-nearest neighbors)
• Hierarchal clustering
• Anomaly detection
• Neural Networks
• Principle Component Analysis
• Independent Component Analysis
• Apriori algorithm
• Singular value decomposition

3.Reinforcement Learning:- It is the ability of an agent to interact with the environment and find out what is the best outcome. It follows the concept of hit and trial method. The agent is rewarded or penalized with a point for a correct or a wrong answer, and on the basis of the positive reward points gained the model trains itself. And again once trained it gets ready to predict the new data presented to it.

It aims at using observations gathered from the interaction with the environment to take actions that would maximize the reward or minimize the risk. Reinforcement learning algorithm (called the agent) continuously learns from the environment in an iterative fashion. In the process, the agent learns from its experiences of the environment until it explores the full range of possible states.

Suppose the Robot in the above figure wants to go home. And in between lots of hurdles are present in the form of fire. Now, if the Robot is using the Reinforcement Learning algorithm to reach home. Then initially, the Robot will make a lot of mistakes. But, slowly it keeps on learning from the mistakes. Moreover, it keeps on improving and commit fewer mistakes. And finally, he will find the shortest path to reach his home. Means, the Robot learns by trying all the possible paths to reach home and then finally chooses the shortest path.

There are many examples of this Reinforcement Machine Learning type. Some of its famous Examples are:

Video Games: Mario and many other games use Reinforcement Machine Learning.

Robotic Applications: Many Robots use Reinforcement Learning Algorithm.