Definition of Machine Learning

Machine Learning is generally defined in two ways,

Arthur Samuel's definition of Machine Learning  - Simple Definition

Machine Learning is a field of study which gives the computers the ability to perform tasks without being explicitly programmed.

This one is pretty old but, easy to understand. It forecasts the future machines capability. Now, you have a question on how the machines will perform tasks without someone's intelligence. The modern definition will give you the answer.


Tom Mitchell's definition of Machine Learning - Modern Definition

A computer program is said to learn from experience E with respect to some class of tasks T and performance measure P, if its performance at tasks in T, as measured by P, improves with experience E.

E - How many times the task is performed by the machine.
P - How well did the machine performed on the task.
T - The definition of the task.

This definition resembles a kid learning a new sport. Consider he starts playing Tennis(T). Initially, he won't be able to hit the ball properly but, gradually after a number of tires(E) he learns to use the racket properly and performs well in the game. (P)

On Next - Classification of Machine Learning
1.Supervised learning 
2.Unsupervised learning.


Comments

Popular posts from this blog

The beginning of Data Quest

Setting up Python Flask server on internet via Port forwarding

Automate Blog Post creation using Blogger APIs and Python

Creation of Orphan Process in Linux using Python

How to remove TWRP from Android Phone - Nubia Z11 miniS