Introduction to Neural network in AI?

4
2584

Aim :-  This article is written for those who are/wants to learn Artificial Intelligence (AI). Below are the learning’s, you will get from this post –

  1. Small example explaining How AI is making world Intelligent.
  2. 5 attributes which will transform Machine into Artificial Intelligent machine.
  3. Learn Neural network.
  4. Pattern Matching.
  5. Comparison between Human brain and Machine brain

Description :-  Artificial Intelligence (AI) freaks. In my last article, I enlighten you with what is all about AI and what are its components. From my point of view, AI have boundaries. In simple words, Sky is the limit. Below we can see a small example which is explaining how AI is making the world more Intelligent.

Point 1 – Example showing How AI is making world Intelligent

Captcha with Artificial Intelligence

  1. Earlier – CAPTCHA (used to stop spam) require users to apperceive and type in static distorted characters.
  2. Now – Designed with AI, next generation CAPTCHAs will predict on simple computer games.

The researchers fixated on a broad form of game like CAPTCHAs, called dynamic game. This will challenge the User to perform a game like task interacting with a series of non-statics images. For example – In a “car parking”, the user is required to identify the car from a set of moving objects and drag-and-drop it to the available “parking” place. Now, you are not required to enter the distorted and complex character. It will be no longer a headache for anyone.

Point 2 – These 5 attributes will transform Machine into Artificial Intelligent machine.

So, before I enter into detailed study of Neural network, I want describe about the central idea of the intelligence which is expected from a machine. Any normal human being uses his 2- 7 % of logical brain. Researchers are working very hard to implement these 5 attributes into the machine. In other terms, the machine that follows the fundamental attributes of a human brain would be called Artificial intelligent machine. Below screenshot shows the required 5 attributes.

5 attributes required to transform Machine into Artificial Intelligent Machine

Arithmetic operations This part of attribute is responsible for calculations. It could be as simple as 2+2 = 4 or it could be a complex too like (((25*25)/5)*7) +49)*17)/15) +55 = ? Just try and see how much time you will take to calculate this expression. I took around 10 minutes to solve this. I know you all are smarter than me.

The basic ideology of the people is that Computer calculates faster than a normal Human. I must say that this assumption is completely wrong. You must be guessing why is it so? I will explain this to you later with my future posts on AI. Also, being a biological greater gene than any other being or any other machine, why we are not able to calculate faster?

FYIShakuntala Devi the first human computer. She can calculate multiplication of 13* 13 Random digits in less than 30 seconds the.

Factual thinking/logical Logical part is responsible to drive conclusions from the inference. You will get it better with this example –If A belongs to B and B belongs to C. So, it might be possible A and C shares some common relations. Let’s check whether you got my point or not. If A is an uncle of B and B is brother of C. The question is how both A and C are related to each other? If your answer is – A is uncle of C, you are an intelligent guy. If not, than you are like and we can hire machines with AI.

FYI – In mathematical terms, this attribute is related with the theory of Associations and transitivity

Learning For now we can understand this as – Brain captures an image of what we observe. This can be in any form i.e. light, sound, touch, etc. Then, brain takes decision in future with reference to our past experiences and observations.

Feeling Smile, sadness, happiness, excitement, guilty, etc.

Non-factual We can describe this attribute as Intuitions and Expectations. For such things you have no idea to put the logic but the consequences. Example – I think India will win the ODI series with England. Even though you are unaware of the ground conditions and match statics.

Neural network Component in AI

Point 3 – Learn Neutral network

Now we can discuss about the first component of AI i.e. Neural network. So game begins now. I will start with neural network and later on I will move forward with the other components – Automatic speech processing, Image processing and Machine learning.

Question – Why neural network?

Answer There are many reasons to give this answer. The basic and foremost reasons are given below.

1)      Understanding some basic principle of Information processing.

Information processing in Brain

 

2)      Study the brain’s problem solving capacity (Studying the application of those basic  principle).

 

Solve problems by Artificial Intelligence

Examples showing Brain Power :-

  • Object recognition even in noisy or partly obscured data.
  • Understanding of context from pieces of information.
  • Learning of language, cycling, social behavior, etc.
  • Face recognition (in 0.1 seconds).

Point 4 –  Associative memory and Pattern recognition

Associative memory An initial pattern “near by” to some stored memory is retrieved, partial input sufficient for memory retrieval

Associative  Pattern matching in Artificial Intelligence

Pattern association When we make computer to learn Associate date then it works with principal of asscoaiatvity

Pattern Association in Artificial Intelligence

Point 5 – Comparison between Human Brain and Machine brain

Comparison between Human Brain and Machine Brain  in Artificial Intelligence

Thanks to all for reading our post on Neural network in AI. Soon I am going to provide small videos with some example of AI. Stay tuned for more.

VN:F [1.9.22_1171]
Your Feedback hels to Improve
Rating: 10.0/10 (5 votes cast)
VN:F [1.9.22_1171]
Rating: +4 (from 4 votes)
Introduction to Neural network in AI?, 10.0 out of 10 based on 5 ratings
  • Wow. You have explained it very well in good manner. Love your way of explaining n writing

    • Abhishek

      Thanx and I also appreciate your work with PhPring Specially the Video part. Good start. 🙂

  • Hi,
    With this post i can co-relate information about neural network with the learnings i got from Biological classes. 🙂
    I think inducing these 5 attributes into a machine is a hard thing but a feasible one.
    Artificial intelligence is giving a great help to the world. Recently I heard that someone from india made a robot who can protect our border. Really awesome – great help to soldiers !!!
    Keep Posting and Sharing AI.
    Thanks,
    Chander Sharma