Deep Learning Introduction

Source: Deep Learning on Medium

Welcome to the course on Deep Learning .When we hear the word deep learning(DL) , lots of thought splashes in our mind about machine learning(ML) and artificial intelligence(AI).We generally think all of these are one and the same. So lets clear it out.

AI : Computer system that mimics or replicates human intelligence .By this definition we can conclude that humans and AI are analogous to each other . For our understandings let us co-relate working of human intelligence and AI.

For human intelligence to be applied we need data(input) , in order to get so we have some abilities like eyes , ears , etc .So lets take an example , we get data from our eyes that , there are 2 living beings in your room.

After getting data our task is to analyse data. 2 living beings , their structure , their color , their neck , their head … etc . So we concluded that there is a giraffe and a dog in your room .But how our brain completed the task ?

Lets say our brain used some methods to complete task . Like it has pre-stored data , giraffe has long neck ; dog has short neck …. etc . so our brain applied intuition as a method to complete task .Now its done as a human part , lets see how AI does .

In above paras three words are highlighted abilities, task , method .As we know human intelligence and AI is analogous . So first step for AI is to get data through its abilities . Computer can gather data by abilities such as computer vision and all . Lets say , we have given a pic of dog and a pic of giraffe and machine’s task is to recognize them . To complete task machine should use some methods .

Those methods can be inn the form of if else statement code . like if tail==short print(‘giraffe’) else print(‘dog’) …. etc and all .This code is written on the basis as pre-stored data analogous to humans . This type of method is know as expert system . But sometimes it becomes very difficult to code each and every parameter . So , human thought why not build a method which can learn by itself upon give input .

This way ML was introduced . ML uses modules like logistic regression , graphical modules to complete task . DL is the family of ML which uses neural networks to complete the task . DL is advance form of it . It deals with large amount of data and learning .Lets see the formal definitions of these terms .

Machine learning is an application of artificial intelligence (AI) that provides systems the ability to automatically learn and improve from experience without being explicitly programmed.

Deep learning (also known as deep structured learning or hierarchical learning) is part of a broader family of machine learning methods based on learning data representations, as opposed to task-specific algorithms.

Artificial intelligence (AI) is an area of computer science that emphasizes the creation of intelligent machines that work and react like humans.

I guess whole of your confusion is cleared . From next blog we will start by python3.x , as it is required for deep learning .

Thanks for joining me!