Types of Machine Learning problems

In the previous blog, we had discussed brief about What is Machine Learning? In this blog, we are going to learn about the types of ML. 

ML is broadly classified into four types:

• Supervised Learning
• Unsupervised Learning
• Semi-supervised Learning
• Reinforcement Learning

1. Supervised Learning

• Supervised learning is where there are input variables, say X and there are corresponding output variables, say Y. We use a particular algorithm to map a function from input(X) to output(Y).
• Mathematically, Y=f(X).
• Majority of the ML models use this type of learning to feed itself and learn.
• The goal of supervised learning is to approximate the said function so well that whenever we enter any new input, it's output is accurately predicted.
• Here, we can say that there is a teacher who guides the model if it generates incorrect results and hence, the machine will keep on learning until it performs to desired results.
• Supervised Learning can be further classified into:
• Classification: Here, the output variable is a category. Example: diseased or not diseased, cooking or not cooking, etc.
• Regression: Here, the output variable is a real value. Example: rupee, dollar, etc.
• Few supervised learning algorithms:
• Linear Regression (For regression)
• Random Forest (For regression and classification)
• Support Vector Machine (For classification) 

Figure 1: Supervised Learning

Figure 2: Classification and Regression in ML

2. Unsupervised Learning

• Unsupervised Learning is where we have input variable(X) but no output variable(Y).
• The goal is to model the underlying structure in data in order to learn more about the data.
• Here, as there is no output pre-determined, there is no teacher and no correct values unlike supervised learning.
• The algorithm is own it's own to discover the correct outputs and learn own it's own.
• Unsupervised learning can be further classified into:
• Clustering: Used when we want to find the group of the date. Example: Group of stocks listed in Nifty50, Group of people who are Corona positive, etc.
• Association: Used when we wan to find rules that describes a large portion of data. Example: People who buys stock X will tend to buy stock Y, People who purchases X will tend to purchase Y, etc.
• Few unsupervised Learning algorithms:
• K-means (For clustering)
• Apriori Algorithm (For association)

Figure 3: Unsupervised Learning

Figure 4: Clustering and Association ML

3. Semi-supervised Learning

• In this type, we have a large collection of input data(X) but few collection of output data(Y).
• This type sits between supervised learning and unsupervised learning.
• Example: Photo gallery, where few are labelled pictures and rest are unlabeled. 
• Labeling data is time-consuming, tedious and expensive while unlabeled data are cheap and easy to get.

Figure 5: Semi-supervised Learning

 4. Reinforcement Learning:

• Here, the program interacts with dynamic environment, in which it plays certain role.
• Program provides feedback of output if it is correct or not and machine learns through it.

Figure 6: Reinforcement Learning

That's all about the types of ML and it's learning.

Hope it was worth investing time!

For further queries related to the blog, you can ask in the comments.