The following series of blogs run in parallel with the Artificial Neural Networks(A.N.N) which enables you to take baby step’s in Machine Learning.

What exactly is Machine Learning ?

Machine Learning is an ability given to a machine to sense, act and react to a particular situation.

In the above picture,the model detects the objects present in the image. Each detected object is represented inside a bounding box.

Types of Machine Learning

Machine Learning is broadly classified into 2 types:

Supervised Learning : In supervised learning, the output is known.In this type of learning, the machine gets trained on the labelled dataset i.e, for every input value given has a specific output. Simply, it can be understood as learning with guidance.

In the above example, the images of different types of cats were given to the machine with the label as “cats” and when a new image is given it classifies them as cats or not.

Supervised learning is again sub-categorized into 2 categories based on label.

• Regression : The output is continuous i.e, output can take many values.
• Classification : The output is discrete i.e, the output will be from a set of fixed number of classes.

Unsupervised Learning : In unsupervised learning, the output is unknown. In this type of learning, the machine gets trained on unlabelled dataset i.e, there is no specified for a given input.Simply , it can be understood as self-learning in which the machine draws it own pattern’s and learns from data without any supervision

In the above example, the machine categorizes them according to their similarities, patterns, and differences.

Types of Machine Learning problems

1. Classification : In classification problem, the output belong to a category like “True” or “False“.
2. Regression Analysis : In regression analysis, we predict a continuous output variable based on one or more predictor variables.
3. Cluster Analysis : Cluster analysis is task of grouping the data points into groups.
4. Anomaly Detection : Anomaly Detection is the technique of identifying rare events which can be suspicious i.e, different from rest of the events occurred.
5. Association Analysis : Association Analysis finds interesting associations and relationships among large set of items.

Conclusion

In this post, we have discussed about the buzzword “Machine Learning”, types of machine learning and machine learning problems. In the next post we’ll learn how to design a solution for a machine learning problem with questions like

• Where to Start?
• How to choose an algorithm?
• How to validate our model?
• How to test our model?
• And so on.

This is something very important to know as a beginner to learn ML in a proper approach. Stay tuned for our next post. Until then, cheers✌️.