Types of Machine Learning

There are mainly three types of Machine Learning:

• Supervised Learning

• Unsupervised Learning

• Reinforcement Learning

Supervised Learning

"I know how to classify this data, I just need you (the classifier) to classify it"

In this form of learning, the right answer/label is already known to the learning algorithm.

It mainly covers two types of problems:

• Classification Problem: The aim is to classify a given object into discrete class labels.

  Ex. Deciding if a person has a disease or not.

• Regression Problem: It deals with problems having continuous values for the output.

  Ex. Estimating the price of a house.

Examples of Supervised Learning: Logistic Regression, Decision Trees, Random Forests, SVMs etc.

Unsupervised Learning

"I have no idea how to classify this data, can you (the algorithm) create a classifier for me?"

In this form of learning, the right answer/label is not known in advance.

We simply have a large data set and the computer program must decide what to do with it.

It encompasses the following types of problems:

• Clustering Problem: In this problem, the algorithm must group the data into clusters containing similar data. Ex. Google News uses clustering algorithms to group similar news articles together.

• Some others include dimensionality reduction and association rule learning: Associative Memory, for example, could be used for a problem where a doctor must diagnose a patient based on symptoms and the past associations those symptoms had with a particular disease.

Examples of Unsupervised Learning: Apriori Algorithm, K-Means Clustering etc.

Reinforcement Learning

"I have no idea how to classify this data, can you classify this data and I'll give you a reward if it's correct or I'll punish you if it's not"

In this form of learning, the machine is trained to make specific decisions by exposing it to an environment where it trains itself continually using trial and error, with the aim of maximizing its reward.

Example of Reinforcement Learning: Markov Decision Process

This course mainly focused on Supervised and Unsupervised Learning techniques.

The next chapter discusses a few Supervised Learning techniques.