Jawaharlal Nehru Technological University Kakinada 2010-2nd Sem M.Tech Computer Science & Engineering Supplementary s Neural Networks (CSE – CS Engineering) - Question Paper

Subject Code: B5806
M.Tech II Semester Supplementary exams June 2010
Neural Networks
(CSE – CS Engineering)
Time: three Hours Max Marks: 60
ans any 5 ques. All ques. carry EQUAL marks


1. a) Consider a multilayer feed forward network, all the neurons of which operate in
their linear regions. Justify the statement that such a network is equivalent to single layer
Feed forward network.
b) explain in detail models of a neuron.
2. a) discuss credit assignment issue in detail.
b) discuss various learning paradigms.
3. a) provide and discuss the relation ranging from the perceptron and bayes classifier for a Gaussian
Environment.
b) What is a perceptron? Differentiate ranging from perceptron representation and perceptron
Training.
4.a) discuss about heuristics for making the Back propagation algorithm perform better.
b) Implement a back propagation network to simulate EX -OR function
5. a) Write about the approximation made in Hessian based pruning techniques.
b) discuss cross-validation and variants of cross-validation.
6. a) explain in detail about the properties of feature map .
b) explain how the "Winner –Take-All " in the Kohonen's layer is implemented and discuss
the architecture and also discuss the training algorithm.
7.a) define about neurodynamical models in detail.
b) define about the manipulation of attractors as a recurrent network paradigm.
8. define the discrete Hopfield model as a content addressable memory and briefly state the
steps involved in the storage phase and retrieval phase of operating a Hopfield network.12m

Earning:   Approval pending.