This page contains the reliably extracted subset, not the complete original paper.
Questions and published explanations below are available without starting a test. Some questions may not have a published solution yet.
0 / 28 published solutions
Question 60 MSQ · 1.0 marks
In the planning graph, which of the following are mutex proposition pairs in Layer 1?
A.
holding(A), holding(C)
B.
holding(A), holding(D)
C.
holding(A), on(C,B)
D.
onTable(A), on(C,B)
E.
on(C,B), on(C,D)
**Deep Learning**
**Section Id :** 64065329294
**Section Number :** 3
**Section type :** Online
**Mandatory or Optional :** Mandatory
**Number of Questions :** 17
**Number of Questions to be attempted :** 17
**Section Marks :** 50
**Display Number Panel :** Yes
**Group All Questions :** No
**Enable Mark as Answered Mark for Review and**
Yes
**Clear Response :**
**Maximum Instruction Time :** 0
A published solution is not available for this question yet.
Suppose that we use Dropout scheme to regularize the deep learning model. Assume that the
dropout probability p = 0.75 for all the layers in the model and the model is trained using
Stochastic Gradient Descent (SGD) optimizer. Choose the correct statements about Dropout
regularization.
A.
Dropout reduces (removes) 25 percent of weights from the model during
training and inference
B.
For each iteration, 75% of neurons are dropped randomly from the model
C.
For each iteration, 25% of neurons are dropped randomly from the model
D.
Each weight gets multiplied by a factor p during inference.
E.
Each weight gets multiplied by a factor 1 − p during inference.
F.
The Dropout method can not be combined with other regularization methods
A published solution is not available for this question yet.
The figure below represents a Boolean function using a unit cube. Each vertex shows the label or
output of the function for the given input. Which of the followings is (are) true about the function?
A.
The function has 8 inputs
B.
The function has 12 inputs
C.
The function has 24 inputs
D.
The function is linearly separable
E.
The function is not linearly separable
A published solution is not available for this question yet.
A team has a data set that contains 1000 samples for training a feed forward neural network.
Suppose they decided to use the gradient descent algorithm to update the weights. Suppose
further that they use line search algorithm for the learning rate as follows, η = [0.01, 0.1]. How
many times do the weights get updated after training the network for 20 epochs? (Note, for each
weight update the loss has to decrease)
A.
20
B.
20000
C.
40000
D.
40
A published solution is not available for this question yet.
Based on the above data, answer the given subquestions.
In the previous question, if we change all the weight values to 10, that is, b = w1 = w2 = · · · = wn =
10, then enter the number of weights that will get updated during the next iteration.
A published solution is not available for this question yet.
An MRI image of size 250×250×8 is feed as an input to a convolutional layer. Suppose that we want
to produce an output of size 119×119×64 using a set of kernel. Suppose the kernel size is 13 × 13.
Answer the given subquestions.
What is the value of stride s?. Assume zero padding (p = 0)
A published solution is not available for this question yet.
An MRI image of size 250×250×8 is feed as an input to a convolutional layer. Suppose that we want
to produce an output of size 119×119×64 using a set of kernel. Suppose the kernel size is 13 × 13.
Answer the given subquestions.
How many parameters (including bias) are there in the convolutional layer?
A published solution is not available for this question yet.
An MRI image of size 250×250×8 is feed as an input to a convolutional layer. Suppose that we want
to produce an output of size 119×119×64 using a set of kernel. Suppose the kernel size is 13 × 13.
Answer the given subquestions.
Suppose that the input image is of size 250 × 250 × 3 × 8. Suppose further the kernel size is 13 ×
13. How many parameters (including bias) are there in the layer now?
A published solution is not available for this question yet.
An MRI image of size 250×250×8 is feed as an input to a convolutional layer. Suppose that we want
to produce an output of size 119×119×64 using a set of kernel. Suppose the kernel size is 13 × 13.
Answer the given subquestions.
The statement that applying max pooling with stride 2 (s = 2) reduces the number of parameters
by half is
A.
TRUE
B.
FALSE
A published solution is not available for this question yet.
Which of the following statements is(are) true about the vanilla RNN ?
A.
It can’t have more than one hidden layer in it
B.
It is prone to vanishing gradient problem
C.
It is prone to exploding gradient problem
D.
Vanishing gradient problem can be solved if ReLU activation is used across the
network
**Sw Testing**
**Section Id :** 64065329295
**Section Number :** 4
**Section type :** Online
**Mandatory or Optional :** Mandatory
**Number of Questions :** 18
**Number of Questions to be attempted :** 18
**Section Marks :** 100
**Display Number Panel :** Yes
**Group All Questions :** No
**Enable Mark as Answered Mark for Review and**
Yes
**Clear Response :**
**Maximum Instruction Time :** 0
A published solution is not available for this question yet.
