Quiz Space

January 2023 term · Deep Learning · BSCS3004

Deep Learning Quiz 1: 26 February 2023, Set QPE3 (January 2023 term)

The IIT Madras BS Deep Learning (Deep Learning) Quiz 1 paper sat on 26 Feb 2023, in the January 2023 term, set QPE3: 14 questions for 47 marks in 120 minutes. Every question is below with its answer. Take it as a timed mock test to be marked, or read it through first.

Questions
14
Marks
47
Duration
120 min
MSQ
1
Numerical
9
MCQ
4

Updated

Official paper: IIT M DEGREE AN4 EXAM QPE4 26 Feb 2023 · No negative marking.

Question 1

+3 marksOne or more correct options

Which of the following statement(s) about MP neuron is (are) true?

Select all that apply.

  1. A

    Given linearly separable data points, it learns the decision rule from the datapoints

  2. B

    The inputs to the neuron are Boolean

  3. C

    The neuron must have inhibitory inputs to implement a Boolean function

  4. D

    The neuron always use a linear decision boundary to separate the data points

Show answer

Correct answers

  • B

    The inputs to the neuron are Boolean

  • D

    The neuron always use a linear decision boundary to separate the data points

Question 2

+6 marksNumerical answer
Show answer

Correct answer: 0.5 (accepted within ±0.4)

Question 3

+3 marksOne correct option
  1. A

    reduces

  2. B

    remains 90o

  3. C

    increases

  4. D

    insufficient information

Show answer

Correct answer

  • B

    remains 90o

Question 4

+2 marksNumerical answer

Based on the above data, answer the given subquestions.

What is the minimum number of neurons required to implement the network?

Show answer

Correct answer: 3

Question 5

+3 marksNumerical answer

Based on the above data, answer the given subquestions.

Suppose we use a hard-threshold as an activation function for the neurons in the first hidden layer and add/subtract the outputs from the first hidden layer to obtain the function. The hard threshold function is defined to be

Suppose we use a hard-threshold as an activation function for the neurons in the first hidden layer and add/subtract the outputs from the first hidden layer to obtain the function. The hard threshold function is defined to be

Show answer

Correct answer: 0

Question 6

+2 marksOne correct option

Based on the above data, answer the given subquestions.

  1. A

    TRUE

  2. B

    FALSE

Show answer

Correct answer

  • A

    TRUE

Question 7

+6 marksNumerical answer

Based on the above data, answer the given subquestions.

Show answer

Correct answer: 101

Question 8

+2 marksNumerical answer

Based on the above data, answer the given subquestions.

What is the total number of parameters (excluding bias) in the network?

Show answer

Correct answer: 15

Question 9

+2 marksNumerical answer

Based on the above data, answer the given subquestions.

The number of paths via which the input x1, that is the first element of x, affects the loss is
Note: Path is a directed edge that connects one neuron with other neuron

Show answer

Correct answer: 9

Question 10

+3 marksNumerical answer

Based on the above data, answer the given subquestions.

Show answer

Correct answer: 0.99 (accepted within ±0.01)

Question 11

+3 marksNumerical answer

Based on the above data, answer the given subquestions.

Show answer

Correct answer: -3.05 (accepted within ±0.05)

Question 12

+5 marksNumerical answer

Based on the above data, answer the given subquestions.

Show answer

Correct answer: 1

Question 13

+5 marksOne correct option
  1. A

    d

  2. B

    b

  3. C

    a

  4. D

    c

Show answer

Correct answer

  • C

    a

Question 14

+2 marksOne correct option

The statement that using vannila (plain) gradient descent algorithm always avoids oscillation around local minimum while converging is

  1. A

    TRUE

  2. B

    FALSE

Show answer

Correct answer

  • B

    FALSE