Skip to content
InsemSep 20232019 patternBE IT

Deep Learning (DL) SPPU Insem Question Paper Sep 2023

All 4 questions from this SPPU insem paper, as printed. Read them here or open the PDF.

Open the PDFMost repeated DL questions
Subject code
414443
Pattern
2019 pattern
Exam
Insem
Session
Sep 2023
Max. marks
30
Time
1 Hour
Semester
VII
Paper code
[6188]-202

Printed for: B.E. (Information Technology) (Insem)

Instructions to the candidates

  1. Answer Q.1 or Q.2, and Q.3 or Q.4.
  2. Neat diagrams must be drawn wherever necessary.
  3. Figures to the right indicate full marks.
  4. Assume suitable data, if necessary.

Questions

  1. Q1

    1. a)

      Explain following terms related to multi-layer feed-forward networks. i) Biases ii) Activation functions

      [5]
    2. b)

      Explain loss function for regression operation.

      [5]
    3. c)

      Define and explain the significance of learning rate of a model?

      [5]

  2. Q2

    1. a)

      Explain loss function for classification operation.

      [5]
    2. b)

      Differentiate among RELU, LRELU and ERELU.

      [5]
    3. c)

      What is regularization? Explain the need for regularization.

      [5]

  3. Q3

    1. a)

      Explain the following hyper parameters for the convolutional layer. i) Filter size ii) Output depth iii) Stride iv) Zero-padding

      [8]
    2. b)

      Explain convolution operation in CNN with a suitable example. Take 5*5 input data, 3*3 kernel data and calculate convoluted features.

      [7]

  4. Q4

    1. a)

      Draw and explain architecture of AlexNet.

      [7]
    2. b)

      This part has equations that don’t show as text. Open the PDF to read it.

      [8]