07A80407 Artificial Neural Networks B.Tech Question Paper : scce.ac.in

Name of the College : SREE CHAITANYA COLLEGE OF ENGINEERING
University : JNTUH
Department : Electronics And Communication Engineering
Subject Code/Name : 07A80407/ARTIFICIAL NEURAL NETWORKS
Year/Sem : IV/II
Website : scce.ac.in
Document Type : Model Question Paper

Download Model/Sample Question Paper : https://www.pdfquestion.in/uploads/scce.ac.in/4942-07A80407-ARTIFICIALNEURALNETWORKS.pdf

SCCE Artificial Neural Networks Question Paper

Code No: 07A8040
R07 Set No. 2
IV B.Tech II Semester Examinations,April/May 2012

Related : Sree Chaitanya College Of Engineering 07A70401 Digital Image Processing B.Tech Question Paper : www.pdfquestion.in/4940.html

Common to Electronics And Telematics, Electronics And Communication Engineering
Time: 3 hours
Max Marks: 80
Answer any FIVE Questions :

April/May 2011

All Questions carry equal marks :
1. (a) Why reset mechanism is essential in ART network? Explain.
(b) Explain training algorithm of ART Network. [8+8]
2. Draw full counter propagation network (Full CPN) architecture and explain the Training phases of the Full CPN. [16]

3. (a) Explain how neural network principles are useful in control applications.
(b) Discuss a neural network model for energy minimization in a texture classification problem. [8+8]

4. (a) Explain the biological prototype of neuron. Also explain the characteristics of neuron.
(b) List and explain the various activation functions used in modeling of artificial neuron. Also explain their suitability with respect to applications. [8+8]

5. Explain the Widrow-Ho learning rule for supervised learning in neural networks with help of an example. Why is it sometimes called the LMS learning rule? [16]
6. (a) Discuss the methods, which have been developed to improve generalization of neural network learning.

(b) Explain the following:
i. Activation function involved in the computation backpropagation
ii. Rate of learning in backpropagation algorithm. [16]

7. Discuss algorithm for storage of conformation in Hopfield network. Explain recall algorithm. [16]
8. Explain the architectures of popular self-organizing maps. Derive the training algorithm of Kohonen network. Also explain how SOMs can be used for data compression. [16]

Code No: 07A80407
R07 Set No. 4
1. Explain the architectures of popular self-organizing maps. Derive the training algorithm of Kohonen network. Also explain how SOMs can be used for data compression. [16]

2. (a) Write history of artificial neural system development.
(b) List and explain the various activation functions. Also explain their suitability with respect to applications. [8+8]

3. (a) Define and explain energy (Lyapunov) function of Hopfield Neural Network.
(b) Discuss storage capacity and energy function of the Hopfield network. [8+8]
4. (a) Derive the weight update equation for discrete Perceptron and write its summary algorithm.
(b) Explain the limitations of backpropagation learning. Also explain the scope to over come these limitations. [8+8]

5. Explain Kohonen’s self-organized feature map algorithm and mention its applications. [16]
6. What are dierent types of learning schemes used in training of artificial neuralnetworks? Explain each of them clearly with suitable examples. [16]

7. Discuss how a particular neural network is selected for a particular problem, viz.,optimization problem, pattern recognition problem and classification problem. [16]
8. (a) Sketch the architecture of Boltzmann network and mention the steps for recall Procedure.
(b) State BAM energy function. [10+6]

April/May 2012

IV B.Tech II Semester Examinations,:
Artificial Neural Networks :
1. (a) Explain the differences between neuro-computing and conventional computers computing.
(b) List and explain the various activation functions used in ANN. [8+8]

2. Bring out the limitations of single layer Perceptrons in computing the logical functions. Suggest an alternative network to overcome the above limitations and explain the suitable learning rule for the suggested network. [16]

3. With a neat architectural diagram explain the application procedure used in Boltz- mann machine. What are limitations of the Boltzmann learning? [10+6]

4. (a) Define the problem of handwritten digit recognition. With suitable diagram, explain architecture of multilayer feed forward network for handwritten character recognition.
(b) Summarize the training algorithm of multi category single layer Perceptron networks. [8+8]

5. With a neat sketch explain operation of Kohorens self-organizing feature map (SOM) algorithm. And explain for what type of problems it is most suitable. [16]

6. (a) What are the assumptions to be satisfied for a network to form a Hopeld network?
(b) Construct an energy function for the same size with N neurons. Show that the energy function decreases every time as the neuron output changes.[8+8]

7. Explain how a feed forward network can be used for character recognition. Use a sample of 710 pixel matrix for the recognition of letter \B”. [16]
8. (a) Explain briefly about the counter propagation trainning algorithm.
(b) Explain the architecture of Grossberg layer and its learning algorithm. [8+8]