SET 1
Code No: 37011
.in
JAWAHARLAL NEHRU TECHNOLOGICAL UNIVERSITY HYDERABAD R05 IV B.Tech. I Semester Supplementary Exams, May/June – 2009 NEURAL NETWORKS AND FUZZY LOGIC (Common to EEE, E.CON.E, MEP, AE, ICE, & AME) Time: 3 hours Max Marks: 80 Answer any FIVE Questions. All Questions carries equal marks. -----
Give a brief account on neural networks. Also, explain what is supervised and unsupervised learning with example. [16]
2.
Discuss, how a recurrent neural network is different from a feed forward neural network? [16]
3.
Explain the classification model, features and decision regions in single layer perception. [16]
4.
”The choice of learning coefficient is a tricky task in back propagation algorithm”. Support your answer. [16]
5.
Explain the following: a] Hetero-associative memory. b] Auto-associative memory.
[8+8]
6.
Write about classical set theory and classical sets.
[16]
7.
Discuss in detail the methods to generate membership functions.
[16]
8.
Write about the following: a] Indirect learning architecture b] Specialized on-line learning control architecture.
[8+8]
Aj
nt
uW
or ld
1.
--OoO--
Code No: 37011
SET 2
.in
JAWAHARLAL NEHRU TECHNOLOGICAL UNIVERSITY HYDERABAD R05 IV B.Tech. I Semester Supplementary Exams, May/June – 2009 NEURAL NETWORKS AND FUZZY LOGIC (Common to EEE, E.CON.E, MEP, AE, ICE, & AME) Time: 3 hours Max Marks: 80 Answer any FIVE Questions. All Questions carries equal marks. -----
“An IF neuron model is considered as a special case of spiking neuron model”. Justify your answer. [16]
2.
A fully connected feedforward network has 10 source nodes, 2 hidden layers, one with 4 neurons and the other with 3 neurons, and a single output neuron. Construct an architectural graph of this network. [16]
3.a] b]
Compare and contrast supervised and unsupervised learning strategies. Distinguish between Batch learning and incremental(stepwise) learning. [8+8] Define the following terms: a] Pattern b] Classes/Categories c] Features and pattern space d] Decision regions and surface. [4+4+4+4]
5.
Explain the following: a] Address-addressable memory b] Content-addressable memory Explain the following terms a] Relation matrix b] Binary relation c] Identity relation d] Universal relation
nt
6.
uW
4.
or ld
1.
[4+4+4+4]
Explain the following: a] Singular value decomposition b] Combs method
Aj
7.
8.a] b]
[8+8]
[8+8]
What is the limitation of the plant inverse identification? What are the limitations of specialized on-line learning control architecture? [8+8] --oOo--
Code No: 37011
SET 3
.in
JAWAHARLAL NEHRU TECHNOLOGICAL UNIVERSITY HYDERABAD R05 IV B.Tech. I Semester Supplementary Exams, May/June – 2009 NEURAL NETWORKS AND FUZZY LOGIC (Common to EEE, E.CON.E, MEP, AE, ICE, & AME) Time: 3 hours Max Marks: 80 Answer any FIVE Questions. All Questions carries equal marks. -----
What are the applications of neural networks?
2.
Briefly explain the Winner take all learning.
3.
Draw the block diagram of a pattern classifier. Explain about discriminator discriminate and dichotomize. Also explain what an exemplar mean. [16] Explain the selection of number of hidden nodes, sigmoid gain, local minima and learning coefficient in back propagation network. [16]
5.
6. 7.
[16]
[8+8]
Given three sets A, B and C. Prove Demurrage’s laws using Venn diagrams. [16] Discuss the following: a] Fuzzy synthesis evaluation b] Fuzzy ordering c] Preferences and consensus d] No transitive ranking [4+4+4+4] Explain Top-down and Bottom-up approach in ANN. Explain types of faults and its diagnosis using ANN.
Aj
nt
8.a] b]
Write short notes on a] Sigmoid gain b] Threshold value in back propagation algorithm.
[16]
uW
4.
or ld
1.
--OoO--
[8+8]
SET 4
Code No: 37011
Explain the following terms: a] Resting potential b] Nernst equation c] Action potential d] Refractory periods e] Chemical synapses
or ld
1.
.in
JAWAHARLAL NEHRU TECHNOLOGICAL UNIVERSITY HYDERABAD R05 IV B.Tech. I Semester Supplementary Exams, May/June – 2009 NEURAL NETWORKS AND FUZZY LOGIC (Common to EEE, E.CON.E, MEP, AE, ICE, & AME) Time: 3 hours Max Marks: 80 Answer any FIVE Questions. All Questions carries equal marks. -----
[3+3+3+4]
Discuss the simplified model of an artificial neuron. What are the three basic elements of a neuronal model?
3.
Illustrate the training and classification of continuous perception with an example. [16]
4.
Illustrate back propagation algorithm with your own training sets, and explain. [16]
5.
Describe the architecture of BAM.
[16]
6.
Discuss the measure of fuzziness and dissonance.
[16]
7.
Discuss the fuzzy rule based system.
[16]
8.
Write a short notes on decision surface.
[16]
Aj
nt
uW
2.a] b]
--oOo--
[8+8]
