1. What is the primary function of a convolutional layer in a CNN?

• a) Reduce the number of parameters

• b) Detect spatial patterns and features

• c) Convert the image to grayscale

• d) Flatten the image into a vector

2. Which CNN layer helps reduce spatial dimensions while preserving key features?

• a) Convolutional layer

• b) Fully connected layer

• c) Pooling layer

• d) Batch normalization layer

3. In an RGB image, how many channels does the input have?

• a) 1

• b) 2

• c) 3

• d) 4

4) Which CNN layer typically appears at the end of the architecture for classification tasks?

• a) Convolutional layer

• b) Pooling layer

• c) Fully connected layer

• d) Dropout layer

5) . How does transposed convolution differ from standard convolution?

• a) It increases the spatial dimensions instead of decreasing them

• b) It replaces convolutional layers with fully connected layers

• c) It does not use filters or kernels

• d) It operates only on grayscale images

6) Which application commonly uses transposed convolutional networks?

• a) Image classification

• b) Image segmentation and generation

• c) Sentiment analysis

• d) Reinforcement learning

7) In which of the following applications is CNN-based color processing commonly used?

• a) Image segmentation and colorization

• b) Sentiment analysis and speech recognition

• c) Reinforcement learning and game AI

• d) Data compression and encryption

8. What is the main advantage of using dilated convolutions in neural networks?

• a) Reducing overfitting

• b) Increasing receptive field without increasing parameters

• c) Reducing computational cost

• d) Improving pooling efficiency

9. In a dilated convolution, what does the dilation rate control?

• a) The number of filters used

• b) The spacing between kernel elements

• c) The stride of the convolution

• d) The activation function used

10) How do dilated convolutions compare to standard convolutions in terms of feature extraction?

• a) They extract features from a larger area without increasing the number of parameters

• b) They are always slower than standard convolutions

• c) They do not perform well with high-resolution images

• d) They require more memory due to larger kernel size.