What Is Artificial Intelligence ?
Artificial intelligence - Imagine you have no idea about any technology and for the very first time you come across this word that is Artificial intelligence . When you read this word there are two things which come in your mind first is Artificial and the second is the Intelligence . Let's break this word and try to find out what this means.
Artificial : Something which is not natural which is made by the humans. It could be anything but here we will consider it as a machine.
Intelligence : Ability to perform tasks or make decisions according to the situation .
So now we can say that Artificial intelligence is a technology in which we train machine to perform tasks that a normal person does. Using Artificial Intelligence we train the machines in such a way that they behave like human.
For example you may have heard about Sophia (first social humanoid Robot ) built in HongKong, Alexa, Siri, Google Assistance and many of them.
Categories :
So till now we understood what is artificial intelligence. Now let's go more deeper in this. We can categorize the artificial intelligence as weak Artificial Intelligence and strong Artificial Intelligence.
Weak Artificial Intelligence : Let's first understand Weak Artificial Intelligence, it is also known as narrow Artificial Intelligence, it designed and trained for a specific task. These tasks could be anything. Have you ever seen the robots in the industries. They are programed for specific purposes only. So the Industrial robots are the weak Artificial Intelligence and as we taled earlier Alexa, Siri they are the weak Artificial Intelligence.
Strong Artificial Intelligence : Now let's come to the second one that is Strong Artificial Intelligence, it is also known as artificial general intelligence (AGI), that is programming which is an image of human brain. These are general purpose Artificial Intelligence, for example if we give them very unfamiliar task, it uses fuzzy logic to apply knowledge and finds a solution. In theory, it is said that a strong Artificial Intelligence should be able to pass a Turing Test as well as the Chinese room test. Eugene Goostman was a first chatbot to pass the Turing Test.
Till now we have seen the definition and categorization of the artificial intelligence. Now let's go to the types of Artificial Intelligence.
There Are 4 Types of Artificial Intelligence
Reactive Artificial Intelligence : It uses the basic Artificial Intelligence principles . It uses algorithms to optimize outputs based on a set of inputs. Reactive Artificial Intelligence is very static, it does not learn from the past experiences . It applies the basic algorithm all the time. Thus, it produces the same output given identical inputs. This type of Artificial Intelligence are most trustworthy and reliable.
Deep Blue was a famous reactive Artificial Intelligence machine, it was designed in 1990s as a chess-playing supercomputer. It defeated grandmaster Gary Kasparov in a chess game. Deep Blue was very ordinary machine which was only capable of identifying the pieces on a chess board . It used the basic principles of the chess knowing how each piece moves and defeated Gary. The computer was not able to identify the future moves by its opponent, it was only playing in the present context.
Google’s AlphaGo is another example of it. Just like Deep BLue , AlphaGo is unable to evaluate future moves but it relies on its own neural network and evaluates the present game. We can say AlphaGo is the optimized version of Deep Blue . AlphaGO defeated Lee Sedol in 2016.
Limited memory Artificial Intelligence: It is learning from the past and using the experience in the present. Unlike reactive, it gives different output for the identical inputs. Autonomatic vehicles, for example, can "read the road" and behave accordingly, even "learning" from its past experience.Limited memory Artificial Intelligence is created when the model is properly trained how to analyze and utilize new data or an Artificial Intelligence environment is built and according to the situations it behaves.
Theory-of-mind Artificial Intelligence : These machines are fully-adaptive and have an extensive ability to learn and retain the past experiences. These types of Artificial Intelligence include advanced chat-bots that could pass the Turing Test example Eugene Goostman. Eugene Goostman fooled pepople believing that the Artificial Intelligence was a human being. While advanced and impressive, these Artificial Intelligence are not self-aware . The concept is based on the psychological understanding that other living things have thoughts and emotions but for the machines there is no emotion.
Self-aware Artificial Intelligence : As per the name suggests, we can say that these machines are aware of their existence. This type of Artificial Intelligence have human-level consciousness and understands its own existence in the world, they are aware of emotional state of others. It would be able to understand what others may need based on not just what they communicate to them but how they communicate it.
Where are they used today?
Artificial Intelligence is incorporated into a variety of different types of technology. Here are five examples:
Machine learning : It is one of the most growing technology in the world . In this the machine learns from the data . On the basis of the data provided to the machine while training is used by the machine to make decisions. THere are three type of it:
1. Supervised learning: Data sets are labeled, for every input there is an output .
Example
input 2 output 4,
input 3 output 9,
input 8 output 64
so what will be the output if input is 9
81 right !!
2. Unsupervised learning: Data sets aren't labeled so it is the responsibility of the machine to find the similarity and act accordingly.
3. Reinforcement learning: Data sets aren't labeled but, after performing an action or several actions, the Artificial Intelligence system is given feedback. For exaple suppose you shown a mango image to the machine and since the data is labeled it will say mango but now if you provide apple image then again it will say mango but you will provide feedback that it is apple so now our machine will be able to detect an apple and mango clearly in future.
Machine vision: This is also known as computer vision You may have noticed that when someone breaks the traffic rules , the copy of report is sent to that person's home. How is it possible this is possible because cameras click the image of number plates of the person violated the rules.
Natural language processing (NLP): This is one of the ost interesting thing about Artificial Intelligence. In this human language is processed by a computer program. For example you may have heard about spam detection, which looks at the subject and text of email and tells if it's a spam. Nowadays we are advancing it more to make the best systems.
Robotics: Since the begining of this article we are talking about the robots so you would have a clear vision about them.
Now let's move on to the advantages and the disadvantages of Artificial intelligence .
Advantages :
- They are good for the specific tasks;
- Less time required as compared to human labour;
- Heavy work can be done just by a computer program;
- There is the consistency in the work;
- They can be programmed as per the demand .
Disadvantages :
- Cost to make these machines are very high;
- Deep technical knowledge is important;
- Less qualified workers to make these Artificial Intelligence machines;
- They are mostly designed for a particular tasks , they are not generalized.
In this blog we have covered what is Artificial intelligence, it's categories, type , uses , advantages and the disadvantages. If you like this article please comment down below. We would love to see your comments...
If you want to write for us please contact us .
