Introduction to Artificial Intelligence

Background:-

Term Intelligence has divided people for years about its meaning. It is very difficult term to define and different people define term intelligence differently. Some people define intelligence as mental ability to learn knowledge and apply knowledge to physical world as well as ability to thinking and reasoning. Other definition includes capability to adapt to new environment, ability to evaluate and thought, learn complex ideas and learn from experiences.

It is more difficult to measure intelligence. The most widely accepted method to measure intelligence is IQ (Intelligence Quotient). People are tested with various sets of questions which requires abilities such as logic, memory, mathematics, verbal capabilities, etc. In educational context, people view academic performance as degree of intelligence which is not necessarily correct. People having ability to gain and analyze knowledge and using knowledge as well as experience to solve problem is superior than commanding large number of facts either memorized or learned from books.

Introduction:-

Artificial Intelligence (AI) is the broad branch of computer science which deals with creating intelligent machine like human and training them to do human task. The main goal of artificial intelligence is to create system that can function intelligently and independently.

Before we start further we must know what behaviors are falls under intelligent? Human beings are superior creature on earth and have superior intelligence. Human everyday task such as talking, recognizing people, translating one language to another, questioning, etc. formal task such as playing chess or Go, proving logic theorem and export task such as designing new engineering drawings, financial analysis, medical analysis are intelligent behaviors.

In 1956 John McCathy coined term “Artificial Intelligence”. He is known as father of AI. He along with his colleague organized the famous Darthmouth Conference in summer 1956. Darthmouth Conference is considered to be base of modern AI field. 

 Conference initial proposal states

“Every aspect of learning or any other feature of intelligence can in principle be so precisely described that a machine can be made to simulate it. An attempt will be made to find how to make machines use language, form abstractions and concepts, solve kinds of problems now reserved for humans, and improve themselves”.

Researchers had classified Artificial Intelligence in two categories on the basis of achieving goals. They are:-

1) Weak AI:-

The principle behind Weak AI is that the machine can be made to have similar intelligence as human. These machines carry out pre-planned task based on some rules and environment set by human.

For example, Human player playing computer chess game. Player can feel as if computer is making impressive moves on its own but actually all possible moves are previously feed by human to computer and computer is mimicking as independent and intelligent body.

2) Strong AI:-

The principle behind Strong AI is that the machine have same mind function that of human in future. These machines perform activities on its own without any help of human. Machine should stop depending on human beings to perform different task and take decision on the stimulus. Strong AI is on initial state and there is a long way to go before such machine are developed. Some people suggest that these kind of machine can’t be developed. We have to wait some decades to see further development in Strong AI field. There is no specific example right now.

Artificial Intelligence (AI) and Machine Learning (ML) are used interchangeably. Sometime these interchange use can confuse reader. They are not same thing. Artificial Intelligence is the broad sense of having smart machines while Machine Learning is the current application on Artificial Intelligence. There are two ways Artificial Intelligence works. They are

Symbolic Learning:-

Symbolic learning was dominant paradigm of AI research from mid 1950s until late 1980s. Symbolic learning suggest that human level AI could be achieved by having programmers handcraft code and set large explicit rules for manipulating knowledge. Symbolic learning is implementation of symbolic reasoning or rules engines or expert system or knowledge graphs. A system is called expert system if it implies human knowledge in a computer to solve problem that ordinarily requires human expertise. Most of the Artificial Intelligence algorithms are expert system.

Machine Learning:-

New approach was introduced to take over symbolic learning i.e. Machine Learning. Machine learning needs to feed lots of data so that machine can learn. Machine learning is the way of transform data. Machine learning requires three things:-

  1. Input data
  2. Example of Expected output
  3. A way to measure whether algorithms is doing a job good or not.

Machine learning models are all about finding appropriate representation for their input data. Technically ML is searching for useful representation of some input data, within predefined space of possibilities using guidance from a feedback signal.

Recent Articles

Tutorial on Image Classification Using Keras

Have you ever thought about how we classify and recognize images and objects around us and if you are interested to...

Machine Learning

Machine learning (ML) is study of computer algorithms that can build different analytical model from data.

Contact Tracing with App Review

Introduction: - Contract tracing is the process of identifying the person who have made contact with an infected individual...

Introduction to Artificial Intelligence

Background:- Term Intelligence has divided people for years about its meaning. It is very difficult term to define...

Related Stories

LEAVE A REPLY

Please enter your comment!
Please enter your name here

Stay on op - Ge the daily news in your inbox