What is Machine Learning

In simple terms, Machine Learning means a Machine that learns.

Machine Learning is a branch of Artificial Intelligence which focus on enabling computers to perform tasks based on data and with minimal human interference.

I should make it clear that the ultimate aim of Machine Learning is build intelligent systems which can learn from data, identify patterns and make decisions to help humans not to replace them. Nothing can replace humans.

Above description and definition of Machine Learning is for non-technical people who doesn’t want to builds career in this Area.

Let’s discuss Machine learning with an intention to build a career out of it.

Before understanding Machine learning we should answer what is learning.

Learning is the ability to improve one’s behavior with experience. It means learning is continuous process of doing a task, evaluating result of that task, improving based on the result, and doing the task again in a better and efficient way.

For example, A marathon runner runs for a couple of miles, evaluates his performance, works on his performance and runs again.

What is Machine Learning:

Machine learning is the study of computer algorithms that learns and builds models from data. These models can be used for prediction, decision making or solving tasks.

A computer algorithm is said to learn from experience E with respect to some class of tasks T and performance measure P if it’s Performance of tasks T as measured by P improves with Experience E.

Difference between Traditional Programming and Machine Learning

In traditional programming a programmer or a user writes a piece of code or program and input data into a computer in order to get a result or Output whereas in Machine learning user has to supply computer with Data and output, the computer then builds a Model or Program to increase the efficiency and accuracy of the task.

Suppose we take an example of a Credit card system, by using traditional programming we can get information like who is paying the bills on time and who is not, but if we apply Machine Learning; we can use data such as income, expenditure, demography etc. to determine whether the person will or will not pay the bill on time.

Architecture of Machine Learning

If we consider Machine learning system as a box, there are two main components in it:

Learner
Reasoner

Learner takes Experience and Background knowledge to build Model; this model can be used by the Reasoner to find a solution of a task provided to it.

Types of Machine Learning

There are mainly three ways in which a Machine learns:

Supervised Learning
Unsupervised Learning
Reinforcement Learning

We will discuss types of Machine Learning in the next vlogs as they are huge topic on their own.