Scribd Logo
Upload

Welcome to Scribd!

  • 21 views

    CST444 Soft Computing, June 2023

    Uploaded by

    ANUMOL C S
    The document appears to be an exam for a course on soft computing. It contains 10 questions in Part A worth 3 marks each and 2 full questions from each of 5 modules worth 14 marks each. The questions cover topics like artificial neurons, perceptrons, backpropagation networks, genetic algorithms, fuzzy sets, neuro-fuzzy systems, and multi-objective optimization.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd

    CST444 Soft Computing, June 2023

    Uploaded by

    ANUMOL C S
    21 views3 pages
    The document appears to be an exam for a course on soft computing. It contains 10 questions in Part A worth 3 marks each and 2 full questions from each of 5 modules worth 14 marks each. The questions cover topics like artificial neurons, perceptrons, backpropagation networks, genetic algorithms, fuzzy sets, neuro-fuzzy systems, and multi-objective optimization.

    Original Title

    CST444 SOFT COMPUTING, JUNE 2023

    Copyright

    © © All Rights Reserved

    Available Formats

    PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    The document appears to be an exam for a course on soft computing. It contains 10 questions in Part A worth 3 marks each and 2 full questions from each of 5 modules worth 14 marks each. The questions cover topics like artificial neurons, perceptrons, backpropagation networks, genetic algorithms, fuzzy sets, neuro-fuzzy systems, and multi-objective optimization.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    21 views3 pages

    CST444 Soft Computing, June 2023

    Uploaded by

    ANUMOL C S
    The document appears to be an exam for a course on soft computing. It contains 10 questions in Part A worth 3 marks each and 2 full questions from each of 5 modules worth 14 marks each. The questions cover topics like artificial neurons, perceptrons, backpropagation networks, genetic algorithms, fuzzy sets, neuro-fuzzy systems, and multi-objective optimization.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    of 3

    B 0400csT444052302

    RegNo.:
    APJ ABDUL KALAM TECHNOLOGICAL
    Eighth Semester B.Tech Degree Regular Examination June 2023 (2019

    Course Code: CST444


    Course Name: SOFT COMPUTING
    Max. Marks: 100 .Duration:3 Hours

    PART A
    Answer all questions, eoch carries 3 marks.
    I Draw a simple Artificial Neuron and discuss the calculation of net input. (3)
    2 neuron.
    Compare and contrast biological neuron and artificial (3)
    J State the testing algorithm used in Perceptron Network. (3)
    4 List the stages involved in Backpropagation Network. 3)
    5 Using your own intuition, plot the Fuzzy membership function for the "Age of (3)
    people".

    6 Let A={(x1,0.5), (x2,0.1), (x3,0.9)} and B:{(xI,0.4), (x2,0.4), (x3,0.5)} (3)


    Find intersection, union and complement of both the fuzzy sets.
    7 Draw the flow chart and explain the steps of Genetic Algorithm. (3)
    8 Explain any 3 mutation techniques with example. (3)
    9 Differentiate between linear and nonlinear Multi Objective Optimization Problem. (3)

    ,10 Explain the processes of tuning in genetic-fuzzy rule-based systems. (3)


    PART B
    Answer any onefull questionfrom eoch module, each carries 14 morks.
    Module I

    I1.. a) Implement ANDNOT function using Mc-Culloch Pitts Neuron. . 8

    b) Define linear separability. Justiff -XOR fuffition is non-linearly separable by a 6


    single decision boundary line.

    OR
    12 a) Calculate the output of the neuron y for the following network using
    - l. binary sigmoidal activation function
    2. bipolar sigmoidal activation function

    Page lof3
    0400csT444052302

    b) Using the Hebb rule, find the weights required to perform the following
    classifications of the given input patterns shown in figure. The ..*,' symbols
    represent the value "l" and empty space indicate "-1". consider..l,'belongs to
    the members of class (so has target value | ) and "0" does not belong to the
    members of class (so has target value -l).
    ++++++
    +++
    ++++++
    (.Jrt 66Ort

    Module ll
    l3 a) Implement AND function with binary inputs and bipolar targets using perceptron
    training algorithm.
    b) Draw the architecture of Back propagation Network and explain the training
    algorithm.

    OR
    14 a) What is Adaline? Draw the model of an Adaline Network. 4
    b) Use Adaline to train OR function withbipolar inputs and targets. Perform 2 epoclls IO
    of training.
    Module lll
    15 a) Using intuition and your own definition of the universe of discourse , plot fuzzy
    membership functions to the following variables:
    Liquid level in the tank
    (a)Very small (b) Smail (c) Empty (d) Full (e) Very tull

    b) Define defuzzification. With the help ofexamples, explain various defuzzification


    methods.

    Page 2of3
    it

    0400csT444052302

    OR
    16 a) consider the discrete fuLzy set defined on the universe X: {q b, c, d, e} as 5
    A= { : *
    T * T.T + 9y, uring zadeh'snotation, find the iu- cut sets for r = l,
    0.9, 0.6, 0+ and 0.
    b) Given two universes X= {xl, x2, x3, x4, x5} and y= {yl, y2,V3,y4, y5}, the g
    fuzzy sets A defined on X and fuzry setB defined on Y are given below.

    1= r#.y.**ff*f,t
    -0.2 0.6 I 0.9 0.7
    B= {i*
    ,t*
    ys* yr* ,}
    i) Find the relation R = A x B
    consider another fuzzy setc defined on the universe V =
    {vl, v2, v3)
    c^ - -0.4 I
    0.8
    {il* a* us}
    ii)FindP=BxC.
    iii) Using max-min composition find Rop.

    Module tV
    17 a) What is the Fuzzy Inference System (FISX lllustrate Mamdani FIS with
    an 7
    example.
    b) Explain the different methods of encoding that are possible in genetic algorithms. 7
    OR
    I 18 a) What is the concept of crossover in Genetic Algorithm? Explain the different 7
    methods of cross over that are possible in genetic algorithms.
    b) Explain any 4 Genetic Algorithm selection operators. 7
    Module V
    n 19 a) Explain convex and non-convex MOOP; 7
    b) Illustrate the different steps in genetic-neuro hybrid systems with the help
    of a 7
    neat block diagram.

    OR
    20 a) What are the properties of dominance relation? 7
    b) What are the classifications of neuro-fuzzy hybndsystems? Discuss in
    detail. 7
    ****

    Page 3of3

    You might also like