| NEURAL NETWORKS AND FUZZY LOGIC |
| PAPER NO. 3 |
 |
 |
 |
|
| PUT ON: Dec, 2003 |
|
EC - 440 NEURAL NETWORKS AND FUZZY LOGIC (B.Tech 8th Semester, 2063) Time : 3 Hours Maximum Marks : 60 NOTE:- This paper consist of Three Sections. Section A is compulsory. Do any Four questions from Section B and any two questions from Section C Section-A Marks : 20 1(a) What is Hebb's rule ? (b) Why training is required for Neural Network ? (c) Defie ART. (d) What can you do with the neural network ? (e) What is role of membership function ? (f) What does 'feedforward' and 'backprop' mean ? (g) Draw diagram of generic neuron and write the function of Dendritics and Axon. (h) What is the advantages of associating weights with inputs ? (i) What is the difference between linear and non-linear activation function ? (j) What do ypu mean by association memory ? Section-B Marks:5 Each 2. What is linear separability problem ? Explain it. 3. Discuss K-mean clustering algorithm. 4. Compare the performance of biological and computer Neural Network. 5. Discuss briefly defuzzification techniques. 6. How recurrent back propagation Neural Network is trained ? Section-C Marks : 10 Each 7. Discuss the ART model of Neural Network. 8. Explain the applications of Neural Networks. 9. Write short notes on the following: (a) Perception Learning Law (b) Widroff Learning Law (c) Hopfield Model. |
| <<PREVIOUS 1 2 3 4 NEXT>> |