What is Machine Learning and How many types are available?
Machine learning is an Artificial Intelligence technology where we give privilege to our computers to access our data and let them use the same data to learn for themselves.
It is basically getting a computer to perform a specific task without being programmed or instructed specifically to do so.
- Machine learning is “learning from a collection of examples on how to perform a particular task”.
- A task is nothing but a small piece of work which a computer/machine may do.
Why learn machine learning?
Machine learning is everywhere around us. Every day, we all use machine learning systems-such as spam filters, recommendation engines, language translation services, chat-bots, personal assistants, search engines, and fraud detection.
Machine learning systems will soon be normally driving our vehicles, and helping doctors diagnose and treat our diseases.
With the advent of machine learning, we can give a computer lots of data to analyze and then that machine can use that data to learn.
Types of machine learning?
There are various ways a machine can learn from the given data sets.
The three main ways of machine learning are:
- Supervised learning.
- Unsupervised learning
- Reinforcement learning
Supervised Machine Learning:
In Supervised learning, The training data or samples that we have collected will be “Labelled” accordingly, It means the data is already tagged with the correct/incorrect answer.
It can be compared with learning taking place in the presence of a supervisor or a teacher.
Unsupervised Machine Learning:
Think of the same examples of coins in supervised machine learning, but instead of labeling the coins according to the weight and dimensions, you will ask the computer to label them accordingly to the overall data after weighing the coins.
Reinforcement Machine Learning:
This type of machine learning is basically reward-based learning, where you will be giving feedback to the computer i.e, either positive or negative.
It varies from other types of supervised learning since the machine is not trained by the sample data set. Instead, it learns from trial and error.
The way humans learn can be given as the best example for reinforcement learning, we have been rewarded when we have done good things and given feedback for our bad doings as well.
( Learning credit goes to Edrobovate, I went through various curriculum’s and developed the content for Edrobovate which is being now taught to 30k+ students across India, Want to learn more about machine learning with the help of their Robotics Courses??
visit: www.edrobovate.com )
by the author.