11. An auto-associative network is:
- A neural network that contains feedback
- A neural network that contains no loops
- A single layer feed-forward neural network with pre-processing
- A neural network that has only one loop
Answer : A Explanation: An auto-associative network is a neural network that contains feedback. So, option A is correct.
12. What kind of neural network is most frequently applied to the image classification?
- RNN (Recurrent Neural Network)
- FNN (Feedforward Neural Network)
- LSTM (Long Short-Term Memory)
- CNN (Convolutional Neural Network)
Answer : D Explanation: CNN (Convolutional Neural Network) is most frequently applied to the image classification. So, option D is correct.
13. What is the primary purpose of canary deployment when deploying a new version of a deep learning model?
- To ensure that only one version of the model is available
- To release the new version to a small subset of users for testing
- To deploy the model in a new location
- To increase the model’s training time
Answer : B Explanation: The Canary deployment releases the new version to a small subset of users for testing before a full rollout and hence, option B is correct.
14. Which of the following is/are Common uses of RNNs?
- Businesses Help securities traders to generate analytic reports
- Detect fraudulent credit-card transaction
- Provide a caption for images
- All of the above
Answer : D Explanation: Common uses of RNNs: Businesses Help securities traders to generate analytic reports, Detect fraudulent credit-card transaction, Provide a caption for images, etc. So, option D is correct.
15. Which of the following statements is true when you use 1×1 convolutions in a CNN?
- It can be used for feature pooling
- It suffers less overfitting due to small kernel size
- It can help in dimensionality reduction
- All of the above
Answer : D Explanation: 1×1 convolutions are called bottleneck structure in CNN. It can be used for feature pooling, it suffers less overfitting due to small kernel size and also it can help in dimensionality reduction. So, option D is correct.
Neural Networks MCQ Questions And Answers
16. Another name for an output attribute:
- Predictive variable
- Estimated variable
- Independent variable
- Dependent variable
Answer : D Explanation: Another name for an output attribute is Dependent variable. So, option D is correct.
17. What is the main benefit of deep neural networks over shallow neural networks?
- Faster training
- Simplicity of architecture
- Ability to learn complex features
- Lower memory usage
Answer : C Explanation: The main benefit of deep neural networks over shallow neural networks is the “Ability to learn complex features”. So, option C is correct.
18. In the context of deep learning model deployment, what does latency refer to?
- The delay between sending a request and receiving a prediction
- The size of the model’s weights
- The time it takes to train the model
- The time it takes to preprocess data
Answer : A Explanation: In the context of deep learning model deployment, Latency refers to the delay between sending a request to a deployed model and receiving a prediction. So, option A is correct.
19. Which of the following is/are the Limitations of deep learning?
- Obtain huge training datasets
- Data labeling
- Both A and B
- None of the above
Answer : C Explanation: The Limitations of deep learning are: Obtain huge training datasets, Data labeling. So, option C is correct.
20. What is an example of value created through the use of Deep Learning?
- Simplifying accountancy by using business rules to create an automated system
- Natural Language Processing (NLP)
- Both A and B
- None of the above
Answer : A Explanation: Simplifying accountancy by using business rules to create an automated system is an example of value created through the use of Deep Learning. So, option A is correct.
