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

- Deep neural network
- Shallow neural network
- Recurrent neural networks
- Feed-forward neural networks

Answer : B

Explanation: Shallow neural network : The Shallow neural network has only one hidden layer between the input and output and hence, option B is correct.

## 22. **Which of the following library comes with a visualization tool for deep learning?**

- Infer.net
- Pandas
- Matplotlib
- None of the above

Answer : A

Explanation: Infer.net is the library that comes with a visualization tool for deep learning. So, option A is correct.

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

- Recurrent neural networks
- Feed-forward neural networks
- Reinforcement Learning
- Convolutional neural networks

Answer : D

Explanation: CNN (Convolutional neural networks) is a multi-layered neural network with a unique architecture designed to extract increasingly complex features of the data at each layer to determine the output. CNNs are well suited for perceptual tasks. So, option D is correct.

## 24. The number of nodes in the input layer is 10 and the hidden layer is 5. The maximum number of connections from the input layer to the hidden layer are :

- More than 50
- Less than 50
- 50
- Can’t say

Answer : C

Explanation: Since MLP is a fully connected directed graph, the number of connections are a multiple of number of nodes in input layer and hidden layer and hence, option C is correct.

## 25. **Which term best defines the process of improving a deep learning model that has already been trained for a particular task using a smaller dataset?**

- Model initialization
- Data augmentation
- Transfer learning
- Gradient descent

Answer : C

Explanation: The process of improving a deep learning model that has already been trained for a particular task using a smaller dataset is called as Transfer Learning. So, option C is correct.

Data Science MCQ Questions And Answers

## 26. **What is the purpose of A/B testing when deploying a deep learning model?**

- To compare the model’s accuracy to a random baseline
- To evaluate the model’s performance in a real-world setting
- To determine the model’s training time
- To train multiple models simultaneously

Answer : B

Explanation: A/B testing is used to evaluate a model’s performance in a real-world setting by comparing it to alternative approaches and hence, option B is correct.

## 27. The input image has been converted into a matrix of size 28 X 28 and a kernel/filter of size 7 X 7 with a stride of 1. What will be the size of the convoluted matrix?

- 22×22
- 28×28
- 19×19
- 21×21

Answer : A

Explanation: The size of the convoluted matrix is given by C=((I-F+2P)/S)+1, where C is the size of the Convoluted matrix, I is the size of the input matrix, F the size of the filter matrix and P the padding applied to the input matrix. Here, P=0, I=28, F=7 and S=1. So, the answer is ((28-7+0)/1)+1=22. So, option A is correct.

## 28. CNN is mostly used when there is an?

- Unstructured data
- Structured data
- Both A and B
- None of the above

Answer : A

Explanation: CNN (Convolutional neural networks) is mostly used when there is an unstructured data set (e.g., images) and the practitioners need to extract information from it. So, option A is correct.

## 29. Deep Learning algorithms are _________ more accurate than machine learning algorithms in image classification.

- 31%
- 37%
- 28%
- 41%

Answer : D

Explanation: Deep Learning can outperform traditional method. For instance, deep learning algorithms are 41% more accurate than machine learning algorithm in image classification, 27% more accurate in facial recognition and 25% in voice recognition and hence, option D is correct.

## 30. **In deep learning, what is the role of a loss function?**

- To normalize input data
- To measure the model’s prediction accuracy
- To initialize model parameters
- To calculate the gradients for optimization

Answer : D

Explanation: In deep learning, the role of a loss function is to calculate the gradients for optimization. So, option D is correct.