Logo 
Search:

Artificial Intelligence Articles

Submit Article
Home » Articles » Artificial Intelligence » ProLogRSS Feeds

Artificial Intelligence sessional paper - 2 of year 2003

Posted By: Milind Mishra     Category: Artificial Intelligence     Views: 2115

Artificial Intelligence sessional paper - 2 of year 2003.

Q1. Give an example and explain in brief. Any Five (10)
1.Alpha-beta cutoffs
2.Waiting for quiescence.
3.Learning in Game Playing scenario.
4.Parallel relaxation in Hopfield networks.
5.Rationality of an agent.
6.The user of an expert system
7.The Unsupervised learning.

Q2. Explain why. Any 10 (10)
1.Perceptron convergence theorem can not extend to multilayer perceptrons.
2.The activation function plays an important role in learning.
3.Boltzman machines are improvement over backpropagation algorithm.
4.It is important that the Game player generate plausible moves.
5.A normal search algorithm does not work for game playing solutions.
6.Iterative deepening produces better results then normal minimax with alpha beta cutoffs in chess like programs.
7.The scope of the expert system must be kept as small as possible.
8.More than one type of explanations to the same situation is needed in Expert Systems.
9.The expert systems must provide explanation to its reasoning in human understandable form.
10.The multi agent scenario is more difficult to handle than single agent.
11.Rationality of an agent does not depend on the outcome of the action of the agent.
12.The agent action in taxi driver case must be goal based.

Q3. Work out. (Any One) (5)
1.How many hidden units are needed for recognizing an image stored using 100*100 pixels. Every pixel is stored using 10 bytes. There are 10 possible images from which we have to recognize. We have a conventional three layer back propagation network. Once finding this out, explain how we achieve the solution if the program for normal backpropagation algorithm is available. 
2.Write a complete prolog program using either list or databases to read and store information about 10 students. Information contains the name of the student, his/her age and a list containing areas of interest. We have to choose a pair of students which together makes the largest set of areas of interest. It may be possible that two student share some common interests. It is also important to have a pair with no more than 5 years gap.

  
Share: 


Didn't find what you were looking for? Find more on Artificial Intelligence sessional paper - 2 of year 2003 Or get search suggestion and latest updates.

Milind Mishra
Milind Mishra author of Artificial Intelligence sessional paper - 2 of year 2003 is from India.
 
View All Articles

 
Please enter your Comment

  • Comment should be atleast 30 Characters.
  • Please put code inside [Code] your code [/Code].

 
No Comment Found, Be the First to post comment!