Professional Documents
Culture Documents
Deep Learning
Assignment- Week 4
TYPE OF QUESTION: MCQ/MSQ
Number of questions: 10 Total mark: 10 X 1 = 10
______________________________________________________________________________
QUESTION 1:
Let X and Y be two features to discriminate between two classes. The values and class labels of
the features are given here under. The minimum number of neuron-layers required to design
the neural network classifier
X Y #Class
1 2 Class-II
0 0 Class-I
-2 -2 Class-I
3 2 Class-II
-1 -1 Class-I
a. 2
b. 1
c. 5
d. 4
Correct Answer: b
Detailed Solution:
Please refer to the lectures of week 4. Plot the feature points. They are linearly separable.
Hence single layer is able to do the classification task.
NPTEL Online Certification Courses
Indian Institute of Technology Kharagpur
QUESTION 2:
Let us assume that we implement an AND function using a single neuron as shown below. The
activation function 𝑓𝑁𝐿 (∙) , of our neuron is denoted as: f(y)=0, for y<30, f(y)=1 for y >=30. What
would be a possible combination of the weights and bias?
a. Bias = 5, w1 = 5, w2 = 25
b. Bias = 10, w1 = 5, w2 = 5
c. Bias = 10, w1 = 15, w2 = 15
d. Bias = 5, w1 = 10, w2 = 10
Correct Answer: c
Detailed Solution:
For AND function, (w1*x1+w2*x2+bias) should be >= 30 only when x1 and x2 equal to 1
and the expression should be less than 30 for all other values of x1 and x2. Only option C
satisfies that. Please refer to the lectures of week 4.
QUESTION 3:
Which among the following options give the range for a logistic function?
a. -1 to 1
b. -1 to 0
c. 0 to 1
d. 0 to infinity
Correct Answer: c
Detailed Solution:
QUESTION 4:
NPTEL Online Certification Courses
Indian Institute of Technology Kharagpur
a. [0.58,0.11, 0.31]
b. [0.43,0.24, 0.33]
c. [0.60,0.10,0.30]
d. [0.67, 0.09,0.24]
Correct Answer: d
Detailed Solution:
𝒙
𝒆 𝒋
SoftMax, 𝝈(𝒙𝒋 ) = ∑𝑛 𝒙𝒌 for j=1,2…,n
𝑘 =1 𝒆
𝒙
𝒆 𝒋
Therefore, 𝝈(𝟑) = ∑𝑛 𝒙𝒌 =0.67and similarly the other values
𝑘 =1 𝒆
QUESTION 5:
Which of the following options is true?
a. In Batch Gradient Descent, a small batch of sample is selected randomly instead
of the whole data set for each iteration.
b. In Batch Gradient Descent, the whole data set is processed together for update
in each iteration.
c. Batch Gradient Descent considers only one sample for updates and has noisier
updates.
d. Batch Gradient Descent produces noisier updates than Stochastic Gradient
Descent
Correct Answer: b
Detailed Solution:
Batch Gradient Descent considers whole dataset for updates in each iteration.
NPTEL Online Certification Courses
Indian Institute of Technology Kharagpur
QUESTION 6:
Choose the correct option:
Correct Answer: b
Detailed Solution:
Follow lecture 17
______________________________________________________________________________
QUESTION 7:
Choose the correct options about the assumptions are generally made during optimization in
machine learning.
Correct Answer: c
Detailed Solution:
NPTEL Online Certification Courses
Indian Institute of Technology Kharagpur
QUESTION 8:
An artificial neuron receives n inputs 𝑥 1 , 𝑥 2 , 𝑥 3 , … . . 𝑥 𝑛 with weights 𝑤1 , 𝑤2 , 𝑤3 , … . . 𝑤𝑛 attached
to the input links. The weighted sum_________________ is computed to be passed on to a
non-linear filter Φ called activation function to release the output. Fill in the blanks by choosing
one option from the following.
a. ∑𝑖 𝑤𝑖
b. ∑𝑖 𝑥 𝑖
c. ∑𝑖 𝑤𝑖 + ∑𝑖 𝑥 𝑖
d. ∑𝑖 𝑤𝑖 𝑥 𝑖
Correct Answer: d
Detailed Solution:
QUESTION 9:
Consider the below neural network. 𝑝̂ is the output after applying the non-linearity function
𝑓𝑁𝐿 (∙) on 𝑦 .The non-linearity 𝑓𝑁𝐿 (∙) is given as a step function i.e.,
0, 𝑖𝑓 𝑣 < 0
𝑓(𝑣) = {
1, 𝑖𝑓 𝑣 ≥ 0
Choose the correct outputs generated by the network when the inputs are
{𝑥 1 = 1, 𝑥 2 = 0, 𝑥 3 = 0} and {𝑥 1 = 0, 𝑥 2 = 1, 𝑥 3 = 1}. Outputs are in the same order as inputs.
a. 1, 1
b. 0, 0
c. 1, 0
d. 0, 1
Correct Answer: c
Detailed Solution:
Detailed Solution:
𝒚 = 𝒙𝟏 𝒘𝟏 + 𝒙𝟐 𝒘𝟐 + 𝒙𝟑 𝒘𝟑 = 𝟐 𝑿 𝟏 − 𝟏. 𝟓 𝑿 𝟎 + 𝟏 𝑿 𝟎 = 𝟐 𝒇(𝒚) = 𝟏 𝒂𝒔 𝒚 ≥ 𝟎
QUESTION 10:
Consider the below neural network. 𝑝̂ is the output after applying the non-linearity function
𝑓𝑁𝐿 (∙) on 𝑦 .The non-linearity 𝑓𝑁𝐿 (∙) is given as a step function i.e.,
0, 𝑖𝑓 𝑣 < 0
𝑓(𝑣) = {
1, 𝑖𝑓 𝑣 ≥ 0
Choose the correct set of weights 𝑤1 , 𝑤2 and bias for which the network behaves as an OR
function.
a. 𝑤1 = 1, 𝑤2 = 1.5, 𝑏𝑖𝑎𝑠 = 1
b. 𝑤1 = 1, 𝑤2 = 0.5, 𝑏𝑖𝑎𝑠 = −1
c. 𝑤1 = 1, 𝑤2 = 1.5, 𝑏𝑖𝑎𝑠 = −1
d. 𝑤1 = 1, 𝑤2 = −0.5, 𝑏𝑖𝑎𝑠 = 1
Correct Answer: c
For OR function
𝒚 = 𝒙𝟏 𝒘𝟏 + 𝒙𝟐 𝒘𝟐 + 𝒃𝒊𝒂𝒔 𝒇(𝒚) = 𝟏 𝒂𝒔 𝒚 ≥ 𝟎, so 𝒙𝟏 𝒘𝟏 + 𝒙𝟐 𝒘𝟐 + 𝒃𝒊𝒂𝒔 should be ≥ 𝟎
for (𝒙𝟏 = 𝟏, 𝒙𝟐 = 𝟏) , (𝒙𝟏 = 𝟏, 𝒙𝟐 = 𝟎), (𝒙𝟏 = 𝟎, 𝒙𝟐 = 𝟏) and should be < 0 for
(𝒙𝟏 = 𝟎, 𝒙𝟐 = 𝟎). Only option C satisfies this condition.
___________________________________________________________________
______________________________________________________________________________
************END*******