Artificial Intelligence

Pooja rajta

2021-11-17T12:47:00.002+05:30

Types of A.I.

2019-12-20T13:15:00.001+05:30

The Artificial Intelligence can be broadly characterised (on the basis of its evolutionary stages) as : -

Artificial Narrow Intelligence
Artificial General intelligence
Artificial Super Intelligence

Artificial Narrow Intelligence

It is as mentioned above the most weakest or the most narrowest stage of Artificial Intelligence, we are on the very edge of the development of this kind of Artificial Intelligence.It includes the self driving cars, AlphaGo,AlphaGo Zero,Siri, Alexa or Google ,etc.

Artificial Narrow Intelligence or Weak A.I. includes Alexa

Artificial General Intelligence

It is the kind of Artificial Intelligence capable of thinking like a human being and improving itself.It is predicted that once we are able to develop such kind of A.I. it would improve itself so exponentially that it will become far more intelligent than a normal human being this is what is called as Artificial Super Intelligence.

Artificial General Intelligence is capable of human like thinking

Artificial Super Intelligence

It is, as discussed above the kind of A.I. very far more intelligent than any human. It will be capable of advancing the technology and knowledge to a level that humans would have took thousands of years to develop in just months or even less. In the end this kind of A.I. is capable of
outsmarting humans.

Artificial Super Intelligence is will be exponentially far intelligent than any human being

A Brief History of A.I.

2019-12-20T13:09:00.002+05:30

A Brief History of Artificial Intelligence.

While in these recent days Artificial Intelligence has gained huge momentum in the world it is first of all worth noting that Artificial Intelligence is not something new.Its birth dates back to the 1950 when Alan Mathison Turing also known as 'the father of artificial intelligence' conducted the so called 'Turing test' which simply posed the question ''Is the machine capable of thinking?'' In this test there are three members two humans of which one is the person who asks some questions and the other is the one who has to answer these questions, and one machine (which needs to be predicted of whether or not is it capable of thinking) who also have to answer the same questions.During the test the questioner has to be isolated and ask the machine and the human these questions and predict from the answers of both the parties, which one is a human.

The idea here is to predict whether the machine is capable of thinking like human or not?(Though many experts now believe that this experiment isn't a good way of judging Artificial Intelligence)

Alan Mathison Turing the 'father of A.I.'

This was how the Artificial Intelligence was first born.

Development in the field of A.I. more formally took its shape after a conference at Dartmouth College in 1956.

But unfortunately the development of this field was stopped between the 70s-80s (A.I. Winter) when the government stopped funding for its research.It again gained momentum in the late 90s which also saw the development of the first computer to defeat a chess champion, Deep Blue created by IBM.

What is Machine Learning?|Definition|Types|Machine Learning Algorithms

2019-12-20T13:05:00.000+05:30

Machine Learning

Machine Learning is a subset of Artificial Intelligence.It is the field of studying scientifically the various ways to give a machine the ability to perform certain operations, though it may not have been explicitly programmed to do so.The machine takes in inputs from surroundings uses certain kinds of algorithms to generate an output based on logical and mathematical calculations and perform accordingly.

Machine Learning to Play Piano

Types of Machine Learning

There are three types of learning in Machine Learning :-

Supervised Learning
Unsupervised Learning
Reinforcement Learning

Supervised Learning

In supervised Learning yo have an input variable X and an output variable Y.To understand this consider an example suppose you wanted to train a robot to differentiate a male face and a female face,you will use the supervised learning criteria to train it.At first you will give the robot labelled data in raw form which can include lots and lots of pictures of man and women labelled s 'Man' and'Women'.The robot uses certain kinds of algorithm, processes it and builds a model that is trained.It then takes input data from its surroundings processes it performs certain calculations based on the algorithms used to train it, and gives output as Man or Women.

Supervised Learning: a model used in Machine Learning

Supervised Learning Algorithms

Linear Regression
Logistic Regression
Decision Tree
Random Forest
KNN
Naive Bayes Classifier

Here is a very short and informal definations of Machine Learning Algorithms

Linear Regression

Linear Regression is used to predict the value of a dependent variable Y with the help of an independent variable Y.

Linear Regression: A Machine Learning Approach

Logistic Regression

Logistic Regression is a statistical method for analysing a dataset in which there are one or more independent variables that determine an outcome which needs to be predicted.Logistics Regression produces the result in a binary format.

Logistic Regression: Another Machine Learning Approach

Decision Tree

A decision Tree is one which has a tree like structure and its internal node represents test on an attribute.Each branch represents an outcome of the test and each leaf node represents the class label.

Decision Tree used in Machine Learning

Random Forest

Random Forest is an ensemble classifier made using different decision trees.In this model we use trees to reduce the risk of overfitting. It produces highly accurate outcomes for large data. It operates by constructing decision trees during training phase. The decision of the the majority of trees is chosen by the Random Forest as the final Decision.

Random Forest: A combination of Decision Trees to provide the best output in Machine Learning

KNN Algorithm

K Nearest Neighbour is a simple algorithm that stores all the available cases and classifies the new data or case based on similarity measure.

KNN Algorithm used in Machine Learning

Naive Bayes Classifier

Naive Bayes Classifier is based on Baye's Theorem which gives the conditional probability of an event A given B.

Bayes Theorem used in Naive Bayes Classifier

Unsupervised Learning

In Unsupervised Learning the data given to the machine is unlabelled.To understand this consider teaching the machine to differentiate between baseballs basketballs and footballs, to do so you would first give the machine, the data which has pictures of baseballs, basketballs and footballs.

The Unsupervised Model would take the data uses its algorithm and processes it to give output that is 'unlabelled'.

Unsupervised Learning

Reinforcement Learning

Reinforcement Learning is one in which an agent or our machine learns how to act in an environment by performing actions in order to get maximum reward. It is more likely based on how we humans think and act.

Reinforcement Learning

Artificial Intelligence(A.I.)|What is Artificial Intelligence?Definition What is Artificial Intelligence?The Rise of A.I.

2019-12-14T13:01:00.001+05:30

Artificial Intelligence a.k.a. A.I. is now one of the most controversial topics today, but What exactly is Artificial intelligence? Here's everything you need to know about A.I.

An A.I. Robot

A Brief History of Artificial Intelligence.

The idea here is to predict whether the machine is capable of thinking like human or not?(Though many experts now believe that this experiment isn't a good way of judging Artificial Intelligence)

Alan Mathison Turing the 'father of A.I.'

This was how the Artificial Intelligence was first born.

Development in the field of A.I. more formally took its shape after a conference at Dartmouth College in 1956.

What is Artificial Intelligence?

So now as we have gained some knowledge of the birth of A.I. its time to actually define it : -
Artificial Intelligence is can be basically defined as 'intelligence created artificially'.
It is about giving a machine the ability to think and act like a human
You can think of it as a kind of machine capable of performing certain logical calculations, coming to a conclusion and acting accordingly(it is so likely how the humans think and act).
You can give a machine the power of thinking and acting by a number of ways one way of doing so is by feeding it with a large amount of labelled data, say you gave a machine lots of pictures of a cat (labelled as Cat) and pictures of dog (labelled as Dog) then by using certain kinds of algorithms you can tell the machine what is a cat and a dog.(This kind of model is used in Machine Learning).
Like any human A.I. learns by collecting and processing data making decisions by using suitable algorithms and performing actions based on certain calculations.

An A.I. Robot Learning to Play Piano

Subsets of Artificial Intelligence

Artificial Intelligence has its subsets mainly as : -

Machine Learning
Deep Learning

Subsets of A.I.

Machine Learning

Machine Learning is the field of study to give machine the ability to live and adapt itself among its surroundings.It is as if to teach the computer to act to something without actually being explicitly programmed o do so.
It provides system the ability to learn and improve from experience.
We can train a machine to do so by giving it data (lots and lots of data) and using suitable algorithms to create a model for it learn and improve from experiences.
The algorithms could be of various types for example : - Linear Regression, Logistics Regression, Decision Tree, etc.

Machine Learning:A subset of Artificial Intelligence

More about A.I.

Deep Learning

Deep Learning is a subset of Machine Learning and its most fascinating thing is that it can 'learn' without being explicitly programmed to do so by considering examples.

This requires the building up of what is called neural networks.These neural networks are similar to your brain(indeed they are actually inspired form human brain).Like any human brain which has billions o neurons that transmit data and processes it to take suitable actions to its surroundings,neural network also works the same way but instead it has three layers input layer, hidden layers and output layer which helps the system to take data, process it by computing various calculations and in the end produce suitable output which helps it to take its 'decisions' and perform actions in proportion and also improve itself.

A Neural Network

More About A.I.

Today we see Artificial Intelligence everywhere from the search engine through which you are reading this page to the A.I. used in big factories in the form of industrial robots and etc.

It is also believed that Artificial Intelligence industry will take a major boom in the following decade.

But it is also surprising that the Artificial Intelligence we use now is just a very fraction of what the potential of A.I. is, indeed if we can create a super intelligent A.I., it is believed that it can open a new world of technology and knowledge.But before discussing these terms lets first understand some more key terms.

Types of A.I.

The Artificial Intelligence can be broadly characterised (on the basis of its evolutionary stages) as : -

Artificial Narrow Intelligence
Artificial General intelligence
Artificial Super Intelligence

Artificial Narrow Intelligence

Artificial Narrow Intelligence or Weak A.I. includes Alexa

Artificial General Intelligence

Artificial Super Intelligence

Conclusion

In the end I would conclude that A.I. can open doors a very advanced page of knowledge but we should remember that

Knowledge and Wisdom does not come hand in hand each time.

Also View How To Earn Money Online