NR
Code No: 55206/MT
M.Tech. – II Semester Regular Examinations, September, 2008 NEURAL NETWORKS (Common to Computer Science & Engineering/ Computer Science/ Software Engineering) Time: 3hours
Max. Marks:60 Answer any FIVE questions All questions carry equal marks ---
1.
Briefly discuss about linear separability and the solution for EXOR problem. Also suggest a network that can solve EX-OR problem.
2.
Explain in detail the differences between competitive learning and differential competitive learning.
3.a) b)
What is meant by perceptron representation problem? Explain why a single layer perceptron cannot be used to solve linearly non separable problems? Give two examples for linearly non-separable problems.
4.a) b)
Write the advantages and disadvantages of perceptron. Explain Least Mean Square (LMS) algorithm.
5. Using back propagation learning, find the new weights for the network shown in figure.1, when presented with an input (0,1) and the target output is 1. Use a learning rate of α = 0.5 and the binary sigmoid activation function.
Figure.1
Contd….2.,
Code No: 55206/MT
::2::
6.
What are the self organizing maps? Explain the architecture and the training algorithm used for Kohonen’s SOMs.
7.
Explain the procedure of identification of dynamical system using neural networks.
8.
What are the modes of operation of a Hopfield network? Explain the algorithm for storage of information in a Hopfield network. Similarly explain the recall algorithm. *****