ML vs AI

 “What is machine learning and is AI (Artificial Intelligence)? Are they are same or different like, water and oil? How are they connected and how are they differentiated?  Why is it so confusing? Why is it so frustrating?”  Have you ever come across those questions when you start to learn Machine Learning and Artificial Intelligence? Let’s see them one by one.

Artificial Intelligent (AI)

The term “Artificial Intelligence” refers to the simulation of human intelligence processes by machines, especially computer systems. In a simple way, Artificial Intelligence refers to “a broader concept of a machine with the ability to perform out the tasks in an extremely methodical way”. It’s unable the machines to think without any human intervention. The machine will make its own decision.

For example Voice recognition, machine vision, Chabot’s, E – payment, Digital assistants, Google map, face detection, gaming, home devices, natural language processing (NLP) and etc.,.

AI programming focuses on three cognitive aspects, such as learning, reasoning, and self–correction. AI can be divided into 3 categories. They are weak AI, General AI, and strong AI.

Machine Learning 

Machine learning is the general term for when computers learn from data. The machine uses a lot of ways like algorithms to learn itself from the data we input. Each ML includes learning and self–correction when it is introduced into new data.

Generally, it provides statistical tools to explore and understand the data. The algorithms can be grouped into 3 categories. They are Supervised, Unsupervised and Reinforcement in addition there is semi-supervised learning.

Supervised learning 

It’s a passed labeled data.  The machine has sample data when we input and output data with labels. It helps us to predict the future with a help of past and present experienced labeled data. It also predicted the errors and self - correct them by using algorithms.

Unsupervised learning 

It has only input data with labels. Output is unknown. A machine may not provide accurate data like supervised data.

Reinforcement learning 

It’s a feedback-based machine learning technique. Machines learn themselves from the feedback given by the users. The feedbacks are either good or bad. The main goal of reinforcement is to increase the positive rewards only.

There is one more learning is called semi-supervised. It works between supervised and unsupervised learning. It contains unlabelled data only.

How Machine learning is different from Artificial Intelligence. Let’s put it together in a table.