1. What is a Convolutional Neural Network (CNN) primarily used for in Computer Vision?

A) Natural Language Processing B) Image and video recognition tasks C) Time series prediction D) Reinforcement learning

2. What is the purpose of the pooling layer in a CNN?

A) To increase the spatial size B) To reduce the spatial size and the number of parameters C) To activate neurons D) To normalize the input data

3. What does "backpropagation" refer to in neural networks?

A) Forward pass of input data through the network B) Process of updating weights by propagating the error backward C) Activation function computation D) Initializing weights in the network

4. What is "overfitting" in the context of neural networks?

A) A situation where the model performs well on training data but poorly on unseen data B) A situation where the model cannot learn the training data C) When the model has too few parameters D) When the model generalizes well to new data

5. Which technique helps prevent overfitting in neural networks?

A) Increasing the number of layers B) Using dropout layers C) Reducing the number of training examples D) Using higher learning rates

6. What is the vanishing gradient problem?

A) When gradients become too large during training B) When gradients become too small, hindering weight updates C) A situation where data is lost during training D) When the model's accuracy suddenly drops

7. What is the purpose of the activation function in a neural network?

A) To introduce non-linearity into the network B) To reduce the size of the network C) To normalize the input data D) To initialize weights

8. What is the primary advantage of using residual connections (as in ResNet) in deep neural networks?

A) They reduce the number of parameters B) They allow gradients to flow more easily during training, mitigating the vanishing gradient problem C) They eliminate the need for activation functions D) They perform data augmentation

9. What is the purpose of data augmentation in Computer Vision?

A) To reduce the size of the dataset B) To artificially increase the size of the training dataset by applying transformations to the images C) To convert images to grayscale D) To normalize the pixel values

10. What does "IoU" stand for in object detection, and what is its purpose?

A) Intersection over Union; to evaluate the overlap between predicted bounding boxes and ground truth B) Input over Units; to measure data throughput C) Iterations over Usage; to track training steps D) Integration over Uncertainty; to assess model confidence