Punjab Technical University 2005 M.Tech Computer Science NEURAL NETWORK AND FUZZY LOGIC CS 501 - Question Paper

NEURAL NETWORK AND FUZZY LOGIC CS 501 second Sem May 2k5

Max marks 100

Note: Attempt any 5 ques.. All ques. carry equal marks.

1. Implement the backpropagation learning algorithm for a fully connected 3 layer network. Include parameters for layer sizes, learning rate, number of training epochs. Test your implementation on the OR problem:

Input vector Target output vector
(0.0,0.0) (0.1)
(0.0,1.0) (0.9)
(1.0,0.0) (0.9)
(1.0,1.0) (0.9)

2. (a) elaborate the features of Hopfield network? How are these features achieved?
(b) What is reinforced learning? explain with example.

3. Explain the subsequent with examples:
(a) Hebb’s Rule
(b) Delta rule

4. (a) discuss the K-means clustering algorithm.
(b) How trend recognition issue can be solved with neural network approach? explain.

5. (a) elaborate the Linguistics variables? How are they used in fuzzy inference?
(b) elaborate the various kinds of membership functions? discuss with examples.

6. (a) elaborate the different steps in the design of a fuzzy system? discuss with examples.
(b) How fuzzy controller can be designed using fuzzy logic?

7. How neural network approach can be applied to speech processing and understanding? elaborate the algorithms which can be used in that? discuss.

8. Write short notes on the following:
(a) Fuzzy if then rules
(b) History of neural network

Earning:   Approval pending.