Machine Learning Basics

Hello folks,    Greetings of the day!!!
Whenever we see machine learning or we hear about machine learning, it seems like magical ones. Lots of questions hit our mind like how this magic happens, how this machines acts such smartly. Well, I am going to take you to the small trip of this magical world. Let’s start with the basic
definition of machine learning.

What machine learning exactly means?

The definition given by Arthur Samuel (1959),

Field of study that gives computers the ability to learn without being explicitly programmed.

Another definition given by Tom Mitchell (1998) based on Well-posed Learning Problem,

A computer program is said to learn from experience E with respect to some task T and some
performance measure P, if its performance on T, as measured by P, improves with experience E.

Got confused right?

Let me elaborate the last definition by taking one example of Classifying calls as spam or not. Here, the task T is to classify the calls as the spam call or not. The experience E is watching you marking numbers as spam calls and Performance P is the numbers of calls correctly identified as spam call or not spam call.

Traditional Programming vs Machine Learning

Going back to first and basic definition of machine learning, this is the block diagram of traditional programming and machine learning.

 Image Source  : Google I/O  ’19

As we can see in the block diagram, in traditional programming we provide the data and rules to the model and according to rules it gives the answer. Whereas in machine learning, we provide the data and answers of that data and based on the provided answer, machine learning model makes rules itself.

Two phases of Machine Learning Algorithm

In machine learning algorithms, actually there are two phases, one is training phase and inference phase. Here are the block diagrams for the corresponding phases.

 Image Source  : Google I/O  ’19

In first phase, we train the model to make the rules such that it can predict results accurately in future and in Inference phase we provide the data and based the self-made rules the machine learning model made the predictions.

The machine learning algorithms are basically branched in two parts, one is Supervised Learning Algorithm and another one is Unsupervised Learning Algorithm.

Supervised Learning Algorithm vs Unsupervised Learning Algorithm

As we have discussed earlier about the two phases, In Supervised Learning Algorithm we provide the data as well as right answers to the Machine Learning Model in Training phase and based on the provided answers it make rules. Whereas in Unsupervised Learning Algorithm, we just provide the data and let the model make the clustering of the data based on having same attributes. Unsupervised Learning Algorithm is also known as Clustering Algorithm.

Let’s take an example to make both terms more clear. Suppose you have an image of one basket filled with different fruits like apples, bananas, grapes, cherry, etc. Now if you apply Supervised Learning Algorithm, you will firstly train model to detect different fruits like apple, banana, cherry and grapes and after you will apply model on input image so it will detects different fruits like this is apple, this is banana and so on.

Now suppose you apply the same example with Unsupervised Learning Algorithm, the model will train itself based on the different attributes like colour, shape, size, etc and when model is applied on input image, it make clusters of fruits having either same colour or same shape and so on.

So concluding this article on introduction to Machine Learning, these were just the baby steps for Machine Learning. There is much more to learn about this magical world of Machine Learning.

Author : Monil Jethva

Monil  is an Electronics & Communications Engineering  final year student at Institute of Technology, Nirma University, Ahmedabad . In initial years, He  did numerous projects on Arduino and Raspberry Pi board like Surveillance on vehicle, Automatic Obstacle Detection on Vehicle, Automatic Traffic Routing, Automatic Railway Gate Control and few more. As of now, He is  working on project Self-Driving Car.  His fields of interests are Machine Learning, Computer Vision and Embedded Systems.  His field of interest are Not only technical aspects, But he has been also a part of Core Committee and managed the National Tech Fest , Cultural fest etc in the college. He likes to help others in learning technical aspects or in case of any problems.

Here are few other interesting topics @Embedkari

How to Learn C  practically ? 

How to do web scraping with Python ? 

How to choose sensor, processing device and network for IoT ?

How Computer Vision is  impacting Processor Road MAP ? 

How device driver is related to Operating System ?

How to Learn Linux Device driver coding ? 

How I can practice AVR, ARM, Linux and other Embedded Questions for Interview ? 

How can I start learning RISC-V without purchasing development board ?

How RISC architecture differs in simple and advanced embedded processors ? 

How to use Home Router to understand Embedded Linux Commands ?

Thanks for reading till end. Please provide your feedback regarding this article. Also subscribe  Embedkari for other interesting topics. .

Embedkari has introduced its first low cost technical training .


Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.