Punjab Technical University 2005 M.Tech Computer Science NEURAL NETWORK AND FUZZY LOGIC CS 518 - Question Paper

NEURAL NETWORK AND FUZZY LOGIC CS 518 second Sem May 2k5

Max marks 100

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

1. What is a Hopfield net? explain the relationship ranging from the stable states of the binary Hopfield net and the graded version of the model?

2. (a) Write down the various form of activation equations for Fy neutron assuming their number as indicated. (please see the attachment)







(b) elaborate the different associative memory models? explain.

3. (a) What is the noise-saturation Dilemma? How is it solved?
(b) What is a Hybrid intelligence system? explain.

4. (a) elaborate the important properties that neuronal signal function should possess?
(b) Show that if delta(x) = 1/(1+exp –lambda x) then delta inverse (.) is monotonic increasing. Plot the inverse signal function.

5. In the situation of neural network, what do you understand by the subsequent terms?
(a) Graceful degradation
(b) Fault tolerance
(c) Cross talk
(d) Model free estimation.

6. (a) Design a perception to separate the half-open convex region shown.(please see the attachment)
(b) How fuzziness is applicable in neural network? explain.



7. (a) What is fuzzy database? What kind of info can be kept in a fuzzy database?
(b) What is a fuzzy control? explain.

8. Write short notes on:
(a) Fuzzy sets
(b) Application of fuzzy logic

Earning:   Approval pending.