Download AKTU B-Tech 6th Sem 2016-2017 NEC013 Artificial Neural Network Question Paper

Download AKTU (Dr. A.P.J. Abdul Kalam Technical University (AKTU), formerly Uttar Pradesh Technical University (UPTU) B-Tech 6th Semester (Sixth Semester) 2016-2017 NEC013 Artificial Neural Network Question Paper

Printed Pageszl Roll No.| l l l l I l l l l l NEC013
B. TECH.
THEORY EXAMINATION (SEM?VI) 2016-17
ARTIFICIAL NEURAL NETWORK
Time : 3 Hours Max. Marks : 100
Note : Be precise in your answer. In case Ofnumerical problem assume data wherever not provided.
SECTION-A
1 Explain the following: (10x2=20)
a) N eural Computing g) N umbers of hidden nodes
b) BNN and AN N h) Pattern association
c) Adaline model i) N etwork inversion
d) ART models j) Components of CL network
e) Boltzmann learning pattern clustering and feature
f) Linear associative network
SECTION-B
2 Attempt any Five of the following: (10x5=50)
a) Draw full counter propagation network (Full CPN) architecture and explain the
Training phases of the Full CPN
b) Explain the biological neuron. Also describe the models of neuron
c) What are the types of learning? Explain the Hebbian learning and Boltzmann learning.
(1) Explain how a pattern classification problems leads to a radial basis function network.
What decides the basic functions in a pattern classification problem?
e) Draw the architecture of MLP network. Derive the expressions used to update weights
in back propagation algorithm for MLP network.
f) Discuss the Recognition of consonant vowel (CV) segments. Explain the texture
classification and segmentation with example.
g) Brie?y explain the Pattern association, Pattern classification and Pattern mapping tasks
of AN N with suitable example.
h) Differentiate between Feed?back neural networks and Feed-forward neural networks.
Explain stochastic networks, simulated annealing and Boltzmann machine.
SECTION-C
Attempt any Two of the following: (15x2=30)
3 a). Discuss algorithm for storage of conformation in Hopfield network. Explain recall
algorithm with suitable example and diagram.
b). Develop simple ANNs to implement the three input AND, OR and XOR functions using
MP neurons. Explain Why XOR problem can?t be solved by a single layer perceptron and how
it is solved by a Multilayer Perceptron.
4 a). 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
b). Explain ART networks and Features, advantages of ART models with suitable example and
diagram.
5 Explain the following with suitable diagram:
a) Principal Component Analysis
b) Vector Quantization
c) Maxican Hat N etworks