Deep Learning | Week 11

Course Name: Deep Learning

Course Link: Click Here

These are NPTEL Deep Learning Week 11 Assignment 11 Answers


Q1. This question has Statement 1 and Statement 2. Of the four choices given after the statements, choose the one that best describes the two statements.
Statement 1: In a max-unpooling layer, the features are placed at the position from where the max-pooled features were taken.
Statement 2: In a max-unpooling layer, features are placed at the left top corner at the time of up sampling.

a. Statement 1is True and Statement 2 is False
b. Statement 1is False and Statement 2 is True
c. Statement 1is True and Statement 2 is True
d. Statement 1is False and Statement 2 is False

Answer: a. Statement 1is True and Statement 2 is False


Q2. This question has Statement 1 and Statement 2. Of the four choices given after the statements, choose the one that best describes the two statements.
Statement 1: In context of triplet loss for face recognition, hard negative example is whose embedding is are far from the anchor, and belongs to same person as the anchor.
Statement 2: In context of triplet loss for face recognition, hard negative example is whose embedding is near to anchor, but belongs to different person other than anchor.

a. Statement 1 is True and Statement 2 is False
b. Statement 1 is False and Statement 2 is True
c. Statement 1 is True and Statement 2 is True
d. Statement 1 is False and Statement 2 is False

Answer: b. Statement 1 is False and Statement 2 is True


These are NPTEL Deep Learning Week 11 Assignment 11 Answers


Q3. How we create the target for Semantic Segmentation?
a. By one-hot encoding the class labels, essentially creating an output channel for each of the possible classes
b. By labelling pixel of each class with different color
c. By creating list of all possible classes with having the index of the pixel belonging to that particular class.
d. None of the above

Answer: a. By one-hot encoding the class labels, essentially creating an output channel for each of the possible classes


Q4. This question has Statement 1 and Statement 2. Of the four choices given after the statements, choose the one that best describes the two statements.
Statement 1: Pixel wise cross entropy Loss can be used as cost function for Semantic Segmentation.
Statement 2: KL Divergence Loss can be used as cost function for Instance Segmentation.

a. Statement 1 is True and Statement 2 is False
b. Statement 1 is False and Statement 2 is True
c. Statement 1 is True and Statement 2 is True
d. Statement 1 is False and Statement 2 is False

Answer: a. Statement 1 is True and Statement 2 is False


These are NPTEL Deep Learning Week 11 Assignment 11 Answers


Q5. This question has Statement 1 and Statement 2. Of the four choices given after the statements, choose the one that best describes the two statements.
Statement 1: Instance segmentation treats multiple objects of the same class as a single entity.
Statement 2: Semantic segmentation treats multiple objects of the same class as distinct individual objects.

a. Statement 1 is True and Statement 2 is False
b. Statement 1 is False and Statement 2 is True
c. Statement 1 is True and Statement 2 is True
d. Statement 1 is False and Statement 2 is False

Answer: d. Statement 1 is False and Statement 2 is False


These are NPTEL Deep Learning Week 11 Assignment 11 Answers


Q6. This question has Statement 1 and Statement 2. Of the four choices given after the statements, choose the one that best describes the two statements.
Statement 1: Semantic Segmentation can be considered as pixel wise classification problem.
Statement 2: Semantic Segmented output has same dimension as the input image dimension.

a. Statement 1 is True and Statement 2 is False
b. Statement 1 is False and Statement 2 is True
c. Statement 1 is True and Statement 2 is True
d. Statement 1 is False and Statement 2 is False

Answer: c. Statement 1 is True and Statement 2 is True


These are NPTEL Deep Learning Week 11 Assignment 11 Answers


Q7. Which of the following can be used as cost function for Semantic Segmentation?
a. Pixel wise cross entropy Loss
b. Mean Square Loss
c. KL Divergence Loss
d. None of the above

Answer: a. Pixel wise cross entropy Loss


These are NPTEL Deep Learning Week 11 Assignment 11 Answers


Q8. Which of the following operation does not increase spatial resolution of features?
a. Max-Unpooling
b. Deconvolution
c. Pixel-shuffle
d. Average-Pooling

Answer: d. Average-Pooling


These are NPTEL Deep Learning Week 11 Assignment 11 Answers


Q9. In a Deep CNN architecture, the feature map before applying a max pool layer with (2×2) kernel is given bellow.

image 22

After few successive convolution layers, the feature map is again up-sampled using Max Un pooling. If the following feature map is present before Max-Unpooling layer, what will be the output of the Max-Unpooling layer?

image 23
image 24

Answer: b


These are NPTEL Deep Learning Week 11 Assignment 11 Answers


Q10. Which of the following operation does not reduce spatial dimension of features?
a. Max-Pooling
b. Conolution with 3 X 3 Kernel, Stride=2, Padding all sides = 1
c. Conolution with 3 X 3 Kernel, Stride=1, Padding all sides = 1
d. Average-Pooling

Answer: c. Conolution with 3 X 3 Kernel, Stride=1, Padding all sides = 1



These are NPTEL Deep Learning Week 11 Assignment 11 Answers

More weeks of Deep Learning: Click Here

More Nptel Courses: https://progiez.com/nptel


These are NPTEL Deep Learning Week 11 Assignment 11 Answers
The content uploaded on this website is for reference purposes only. Please do it yourself first.
Home
Account
Cart
Search