Question Paper of "NEURAL NETWORKS & FUZZY LOGICS" for M. Tech (ECE) (Sem.-1st), SUBJECT CODE: EC-505, Paper ID [ E0497], PTU Question Paper.

ptupaper.com

Roll No……

Total No. of Questions: 08

M. Tech (ECE) (Sem.-1^st)

NEURAL NETWORKS AND FUZZY LOGIC

Subject Code: EC-505

Paper ID: [E0497]

Time: 3 Hrs.                                                                 Max. Marks: 100

Instruction to Candidates:

1.      Attemptany FIVE questions.

2.      All questions carry EQUAL marks.

Q1. (a) Explain architecture of multilayered feed forward network.

(12)

(b) Explain the architecture of radial basis function network.

(8)

Q2. Draw the architecture of Discrete Hopfield network and explain in detail with training algorithm.

(20)

Q3. (a) Write any four properties of Fuzzy sets. Also find the Fuzzy Cartesian Product for the fuzzy sets A={(0.3/x₁) +(0.4/x₂)+ (0.6/x₃) and B={(0.2/y₁)+ (0.1/y₂)+ (0.4/y₃)}.

(b) Describe in detail about fuzzy relations.

Q4. (a) Write step by step training algorithm of back propagation network and also mention its advantages, disadvantages and applications.

(14)

(b) What is the problem of non-linear separability? Explain how it can be  overcome?

(6)

Q5. Consider two fuzzy relations R₁and R₂ on X x Y and Y x Z respectively, where X={a,b,c}, Y={1,2,3,4} and Z=(a,b). Assume R₁and R₂ can be expressed as the following relation matrices

(b) Explain the properties of fuzzy sets.

(20)

Q6. With the help of examples explain various defuzzification techniques.

(20)

Q7. Explain applications of neural networks in pattern recognition and speech & decision making.

(20)

Q8. Explain the following:

(a)    Antilock Braking system

(b)   K-means clustering algorithms

(20)