Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                
Scribd Logo
Upload

Welcome to Scribd!

  • 35 views

    Applied Soft Computing Test1 April2021

    Uploaded by

    Rajendiran S
    The document is a test for a course on applied soft computing. It contains two parts - Part A with 5 short answer questions worth 2 marks each on topics like activation functions, supervised vs unsupervised learning, and when error backpropagation stops. Part B has 3 long answer questions worth 5 marks each, including questions on delta rule, unsupervised learning rule, and an algorithm for backpropagation neural network with a diagram.

    Copyright:

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

    Applied Soft Computing Test1 April2021

    Uploaded by

    Rajendiran S
    35 views2 pages
    The document is a test for a course on applied soft computing. It contains two parts - Part A with 5 short answer questions worth 2 marks each on topics like activation functions, supervised vs unsupervised learning, and when error backpropagation stops. Part B has 3 long answer questions worth 5 marks each, including questions on delta rule, unsupervised learning rule, and an algorithm for backpropagation neural network with a diagram.

    Original Title

    Applied soft computing test1 April2021

    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 is a test for a course on applied soft computing. It contains two parts - Part A with 5 short answer questions worth 2 marks each on topics like activation functions, supervised vs unsupervised learning, and when error backpropagation stops. Part B has 3 long answer questions worth 5 marks each, including questions on delta rule, unsupervised learning rule, and an algorithm for backpropagation neural network with a diagram.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    35 views2 pages

    Applied Soft Computing Test1 April2021

    Uploaded by

    Rajendiran S
    The document is a test for a course on applied soft computing. It contains two parts - Part A with 5 short answer questions worth 2 marks each on topics like activation functions, supervised vs unsupervised learning, and when error backpropagation stops. Part B has 3 long answer questions worth 5 marks each, including questions on delta rule, unsupervised learning rule, and an algorithm for backpropagation neural network with a diagram.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    of 2

    Department of Electronics and instrumentation engineering

    Puducherry Technological University


    Applied soft computing
    Test 1 26.04.2021 from 2.00 to 2.50 pm
    Part A (5X2=10 Marks )
    Answer all questions
    1. State the bipolar and unipolar continues activation function with the
    significant of λ
    2. Distinguish between the supervised and unsupervised learning
    3. When the error back propagation stops learning?
    4. What are the steps involved in adaptive resonance theory?
    5. How the weights are adjusted in winner take all learning rule?
    Part B (3X5=15 Marks )
    Answer any three questions
    6. Find the updated weights W4 of the neuron which learns based on the
    delta learning rule
    With the inputs of X1 = [2 4 0.5 -1 ] ; X2 = [1 2 0.4 0.5 ]; X3 = [12 1 -0.5 1 ]
    and initial weights are [ 1 2 3 4 ] and the desired outputs are D=[0.6 0.8 -
    0.2]
    7. Find the updated weights W4 of the neuron which learns based on the any
    one type of unsupervised learning rule
    With the inputs of X1 = [2 4 0.5 -1 ] ; X2 = [1 2 0.4 0.5 ]; X3 = [12 1 -0.5 1 ]
    and initial weights are [ 1 2 3 4 ]
    8. Write an algorithm of error back propagation of neural network in detail
    with neural network diagram
    9. Design the neural network for the following truth table
    Input1 Input 2 output
    -1 -1 -1
    -1 1 1
    1 -1 1
    1 1 -1

    You might also like