DEEP LEARNING MCQ

1. Which of the following is a subset of machine learning?

A. Numpy B. SciPy C. Deep Learning D. All of the above

2. How many layers Deep learning algorithms are constructed?

A. 2 B. 3 C. 4 D. 5

3. The first layer is called the?

A. inner layer B. outer layer C. hidden layer D. None of the above

4. RNNs stands for?

A. Receives neural networks B. Report neural networks C. Recording neural networksD. Recurrent
neural networks

5. Which of the following is/are Common uses of RNNs?

A. BusinessesHelp securities traders to generate analytic reports

B. Detect fraudulent credit-card transaction C. Provide a caption for images D. All

of the above

6. Which of the following is well suited for perceptual tasks?

A. Feed-forward neural networks B. Recurrent neural networksC. Convolutional neural networks

D. Reinforcement Learning

7. CNN is mostly used when there is an?

A. structured data B. unstructured data C. Both A and B D. None of the above

8. Which neural network has only one hidden layer between the input and output?

A. Shallow neural network B. Deep neural network C. Feed-forward neural networks D.

Recurrent neural networks

9. Which of the following is/are Limitations of deep learning?

A. Data labeling B. Obtain huge training datasets C. Both A and B

D. None of the above
10. Deep learning algorithms are _______ more accurate than machine learning algorithm in image
classification.

A. 33% B. 37% C. 40% D. 41%

11. Which of the following is not a boundary descriptor?

A. histogram B. Fourior transformation C .polygonal representation D. signature
12. We use gray co-occurance matrix to extract which type of information?

A. Boundary B. Texture [Link] D. Zero crossing rate

13. if the large part of gray co-occurance matrix are connected around the main diagonal, then which one of
the following will be true?

a. the value of element difference moment will be low

b. the value inverse element difference moment will be low
c. the value of entropy will be very low
d. none of the above
14. Which is the following is true about neurons?
A. A neuron has a single input and only single output
B. A neuron has multiple inputs and multiple outputs
C. A neuron has a single input and multiple outputs
D. All of the above
15. Which of the following is an example of deep learning?
A. Self-driving cars B. Pattern recognition C. Natural language processing D. All of the above
16. Which of the following statement is not correct?
A. Neural networks mimic the human brain B. It can only work for a single input and a single output
C. It can be used in image processing D. None
17. Autoencoder is an example of-
A. Deep learning B. Machine learning C. Data mining D. None
Which of the following deep learning models uses back propagation?
A. Convolutional Neural Network B. Multilayer Perceptron Network
C. Recurrent Neural Network D. All of the above
18. Which of the following steps can be taken to prevent overfitting in a
neural network?
A. Dropout of neurons B. Early stopping
C. Batch normalization D. All of the above
19. Neural networks can be used in-
A. Regression problems B. Classification problems C. Clustering problems D. All of the above

20. In a classification problem, which of the following activation function is most widely used in the output
layer of neural networks?
A. Sigmoid function B. Hyperbolic function C. Rectifier function D. All of the above

21. Which of the following is a deep learning library?

A. Tensorflow B. Keras C. PyTorch D. All of the above
22. Which of the following is true about bias?
A. Bias is inherent in any predictive model B. Bias impacts the output of the neurons
C. Both A and B D. None
23. What is the purpose of a loss function?
 A. Calculate the error value of the forward network
B. Optimize the error values according to the error rate
C. Both A and B
D. None
24. Which of the following is a loss function?
A. Sigmoid function B. Cross entropy C. ReLu D. All of the above
25. Which of the following loss function is used in regression?
A. Logarithmic loss B. Cross entropy C. Mean squared error D. None
26. Suppose you have a dataset from where you have to predict three classes. Then which of the following
configuration you should use in the output layer?
A. Activation function = softmax, loss function = cross entropy
B. Activation function = sigmoid, loss function = cross entropy
C. Activation function = softmax, loss function = mean squared error
D. Activation function = sigmoid, loss function = mean squared error
27. What is gradient descent?
A. Activation function B. Loss function C. Optimization algorithm D. None
28. What does a gradient descent algorithm do?
A. Tries to find the parameters of a model that minimizes the cost function
B. Adjusts the weights at the input layers
C. Both A and B
D. None
29. Which of the following activation function can not be used in the output layer of an image classification
model?
A. ReLu B. Softmax C. Sigmoid D. None
30. For a binary classification problem, which of the following activation function is used?
A. ReLu B. Softmax C. Sigmoid D. None
31. Which of the following makes a neural network non-linear?
A. Convolution function B. Batch gradient descent C. Rectified linear unit D. All of the
above
32. In a neural network, which of the following causes the loss not to decrease faster?
A. Stuck at a local minima B. High regularization parameter
C. Slow learning rate D. All of the above
33. For an image classification task, which of the following deep learning algorithm is best suited?
A. Recurrent Neural Network B. Multi-Layer Perceptron
C. Convolution Neural Network D. All of the above
34. Suppose the number of nodes in the input layer is 5 and the hidden layer is 10. The maximum number of
connections from the input layer to the hidden layer would be-
A. More than 50 B. Less than 50 C. 50 D. None
35. Which of the following is true about dropout?
A. Applied in the hidden layer nodes B. Applied in the output layer nodes
C. Both A and B D. None
36. Which of the following is a correct order for the Convolutional Neural Network operation?
A. Convolution -> max pooling -> flattening -> full connect
B. Max pooling -> convolution -> flattening -> full connection
C. Flattening -> max pooling -> convolution -> full connection
D. None
37. Convolutional Neural Network is used in-
A. Image classification B. Text classification
C. Computer vision D. All of the above
38. Which of the following neural network model has a shared weight structure?
A. Recurrent Neural Network B. Convolution Neural Network C. Both A and B
D. None
39. LSTM is a variation of-
A. Convolutional Neural Network B. Recurrent Neural Network
C. Multi Layer Perceptron Network D. None
40. Which of the following neural networks is the best for machine translation?
A. 1D Convolutional Neural Network B. 2D Convolutional Neural Network
C. Recurrent Neural Network D. None
41. Which of the following neural networks has a memory?
A. 1D CNN B. 2D CNN C. LSTM D. None
42. Batch normalization helps to prevent-
A. activation functions to become too high or low B. the training speed to become too slow
C. Both A and B D. None

43. Suppose you are solving a n-class problem, how many discriminant function you will need for solving

A. n-1 B. n C. n+1 D. n-2

44. What will be the nature of the decision surface when the covariance matrices of different classes are
identical but otherwise arbitrary?

A. Always orthogonal to two surfaces

B. Generally not orthogonal
C. Bisector of the line joining two mean but not always orthogonal to two surfaces
D. Arbritrary
45. For minimum distance classifier which of the following must be satisfied?
A. all the classes should be identical covariance matrix and diagonall matrix
B. all the classes should have identical covariance matrix but otherwise arbitrary
C. all the classes should have equal class probability
D. none of the above
46. What is the direction of weight vector w.r.t. decision surface for linear classifier?
A. parallel B. normal C. at and inclination of 45 D. Arbitrary
47. Find the distance point of the 3D point, P=(-3,1,3) from the plane defined by 2x + 2y + 5z + 9 = 0
A. 3.1 B. 4.6 C. 0 D. infinity
48. What is the shape of the loss landscape during optimization of SVM?
A. linear B. Paraboloid C. ellipsoidal D. non convex
49. How many local minimum can be encountered while solving the optimization for maximizing margin for SVM?
A. 1 B. 2 C. INFINITY D. 0

50. Which of the following classifier can be replaced by a linerar SVM?

A. logistic regression B. Neural networks
C. Decision tree D. None of the above
51. For a 2-class problem what is the minimum possible number of support vectors,assume there are more
than 4 examples from each class?
A. 4 B. 1 C. 2 D. 8

52. What will happen to the margin length of a max-margin linear SVM if one of non-support vector training
example is removed
A. Margin will be scaled down by the magnitude of that vector
B. Margin will be scaled up by the magnitude of that vector
 C. Margin will be unaltered
D. Cannot be determined from the information provided
53. Which logic function can be performed using a 2-layer neural network?
A. AND [Link] [Link] [Link]
54. Which of 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
55. The number of weights to be learned by the neural network having 3 inputs and 2 classes and a hidden layer with 5
neurons is
A. 12 B. 15 C. 25 D. 32
56. Which activation function is more prone to vanishing gradient problem?
A. ReLu B. Tanh C. sigmoid D. Threshold
57. Which one of the following activation function is NOT analytically differentiable for all real values of the given
input?

A. Sigmoid B. Tanh C. ReLu D. none of the above

58. PCA reduces dimensions by finding a few_____________________
A. Hexagonal linear combination
B. Orthogonal linear combination
C. Octagonal linear combination
D. Pentagonal linear combination
59. which of the following two vectors can form the first two principal components?
A. {2;3;1} and {3;1;-9}
B.{2;4;1} and {-2;1;-8}
C. {2;3;1} and {-3;1;-9}
D. {2;3;-1} and {3;1;-9}
60. Which of the following autoencoder methods uses corrupted versions of the input?
A. Over complete design B. Undercomplete design
C. Sparse design D. Denoising design
ANSWERS

1. C 11. A 21. A 31. C 41. C 51. C

2. B 12. B 22. D 32. C 42. C 52. C
3. C 13. B 23. C 33. D 43. B 53. D
4. D 14. D 24. C 34. C 44. C 54. C
5. D 15. D 25. B 35. C 45. C 55. D
6. C 16. B 26. C 36. A 46. B 56. B
7. C 17. A 27. A 37. A 47. B 57. C
8. D 18. C 28. C 38. D 48. A 58. B
9. C 19. D 29. C 39. C 49. A 59. A
60. D
10. D 20. D 30. A 40. B 50. A