Answer any four from the remaining
Answer all parts of any Question at one place.
Time: 3 Hrs. Max. Marks: 70
1. write briefly
a. Stochastic Equilibrium b. Hop field Circuit
c. Max – Mini Fuzzy composition d. Subset hood
e. Bi-directional stability
2. a Describe Neural and Fuzzy systems as model free function estimators .
b. Discuss the Taxonomy of nearest Network models
3. a. Discuss commonly used signal functions to model the activation of neural in
neural networks
b. Find the optimal layer associative memory (OLAM) matrix M for the
association given below
A1 = (1 2 3) B1 = (4 3 2 )
A2 = (2 3 4) B2 = (3 5 2 )
A3 = (3 4 6) B3 = (2 2 1)
Determining whether Ai= M - Bi
4. Discuss in detail the two additive bivalent neural Network models and compare
them.
5. a. What is competitive learning? How does it differ from signal Hebbrian
learning.
b. What are Fuzzy Cognitive Maps ? How can they be used to combined
opinion of multiple experts ?
6. a. Discuss the essential differences between supervised and unsupervised learning
in Neural Nets.
b. Discuss various learning algorithms and write their limitations.
7. a. State the fuzzy entropy Theorem and explain it suitable example.
b Write the entropy – subsethood theorem and its implications
8. a. What is Fuzzy centroid defuzzification scheme? Explain how is it used in
FAM system architecture.
b. Use Correlation – Minimum encoding to construct the FAM matrix M from the
fit- Vector pair (A,B) if A=(0.6, 1,0.2,0.9) and B=(0.8 0.3 1.0) . Is (A,B) a bidirectional
fined points? Pass A"=(0.2 0.9 0.3 0.2) through M and B"=(0.9 .5
1.0) through MT . Do the recalled fuzzy sets differ from B and A?
RELATED LINKS
No comments:
Post a Comment