Machine Learning, AI, Deep Learning, and Data Science

Machine Learning, AI, Deep Learning, and Data Science: In this tutorial, we will learn what are Machine Learning, AI, Deep Learning, and Data Science with the help of examples? By Atul Anand Last updated : April 12, 2023

Overview

Well, this is the very first article on Machine Learning and its brother subjects. Before we proceed into studying further, it must be noted that all of these are often considered to be same; especially ML, AI, and Deep Learning. One is the subset of another. You will get better insight going through the article. I have also included some interesting facts in this article. Go through each and every article, and this will become a heavenly place for ML/AI enthusiast. The best thing about this article series is that this is the simplest demonstration of this huge subject through my Learning Experience. Be frank in clearing your doubts in the comment section.

What is Artificial Intelligence?

A field of study constituting set of techniques that try to solve different kinds of problems.
Intelligence displayed by machines, in contrast with the natural intelligence by humans and other animals.
It emphasizes on creation of intelligent machines that work and reacts like humans.

What is Machine Learning?

ML is a subfield of Artificial Intelligence where machines can learn from “DATA” without explicitly Programmed instructions, without rules.
It is the field of making computers (machines) smarter and able to learn from the DATA rather than static instructions.
Large features suggested/ fed manually for learning.

What is Deep Learning?

A subfield of Machine Learning.
It is the set of techniques that help to parameterize neural networks with many, many layers and parameters.
It mainly uses the Deep Neural Network.
It requires huge amount of DATA and Computational Power.
Creates ‘Features’ automatically.

What is Data Science?

It is also known as the "Data- Driven Science".
It is an interdisciplinary field about scientific methods, processes and systems to extract knowledge or insights from DATA in various forms; either structured or unstructured, similar to data mining.
A “Concept to Unify statistics, data Analysis and their related methods” in order to “understand and analyze actual phenomena” with Data.

Evolution of technologies

Example to understand Machine Learning v/s Deep Learning

ML:

House_area * 0.4 + rooms * 0.2 + area_per_room * 0.3 = value_of_house

DL:

House_area * 0.2 + rooms * 0.4 = room_ratio
House_area * 0.3 + room_ratio * 0.3 = value_of_house

Note: Please pay attention here that we needed to feed more parameters( here, called as features in ML) but less layers in Machine learning case; whereas we needed more layers( deep neural networks) in deep Learning case, and more features are automatically generated and used by these neural networks.

Venn diagram

This Venn diagram gives you a deep insight into the thin interfaces between these vital fields of studies.

* Please note that, I am considering all three fields; viz. ML, AI, and DL when I am stating Deep Learning. There’s a reason behind this, it is the modern and latest field of Interest.

Application of Deep Learning

Face Recognition
Voice Recognition
Video Recognition
Text Analytics
Natural Language Processing (since, Large Feature Required)
Robotics (In Automation)

Famous Deep Learning Researchers

Andrew Ng (famous instructor at CourseEra)
Geoff Hinton
Yann LeCun
Yoshua Bengio
Andrej Karpathy

Deep Learning in Big Companies

Google: Buying Deep Mine for $400 Million. Self Driving Car.
Apple: Self Driving Car.
Nvidia: GPU’s innovation.
Toyota: AI research investment of $1 Billion.
Facebook: Face Recognition.

Conclusion About Machine Learning, AI, Deep Learning, and Data Science

As a conclusion, we can say that these three fields can be considered similar despite its minute differences in its contents. But, Data Science is quite different. It includes few techniques of Machine learning and Artificial Intelligence, but not Deep Learning. Altogether, it has some specific techniques. Besides, we can also see that Deep learning is taking a boon in modern world. There is a vast scope in present days and in future for professionals and researchers practicing in one of these fields; especially deep learning if we talk about modern techniques. You will see a progressive content in upcoming articles, which will lead you from very beginners to advance levels of machine learning. Go through important theory parts. Practice lot mathematics involved. Try out the code given, and do alterations. And, make this teaching learning process active (bi-directional), to enhance the learning experience. Catch you later in the next article. HAPPY LEARNING!

How to Learn Machine Learning and Artificial Intelligence?