NNDL

2. What type of neural network is used for unsupervised learning and aims to encode input data into a lower-dimensional representation?
6. What optimization algorithm combines the benefits of both momentum and RMSprop?
11. What approach involves using a pre-trained model as the starting point for a new model?
13. What is the technique for representing categorical variables as binary vectors?
15. What technique is used to assess a model's performance by splitting the dataset into multiple subsets?
17. What is the representation of categorical variables as continuous vectors in neural networks?
19. What is the technique for initializing neural network weights to prevent vanishing or exploding gradients?
20. What type of neural network is designed for processing structured grid data, such as images?
21. What metric combines precision and recall for evaluating the performance of a classification model?

1. The activation function commonly used in the output layer for multiclass classification.
3. What downsampling technique in convolutional neural networks selects the maximum value from a set of values?
4. What is the phenomenon where the gradient becomes extremely small during training?
5. What is the process of artificially increasing the size of a dataset by applying transformations to the existing data?
7. What technique is used to prevent overfitting by adding a penalty term to the loss function based on the magnitude of weights?
8. What is the iterative optimization algorithm used to minimize the loss function?
9. What is the activation function that replaces all negative values in the input with zero?
10. What technique involves normalizing inputs in a neural network to improve training stability?
12. What is the process of finding the best hyperparameter values for a machine learning model?
14. What is the technique of gradually reducing the learning rate during training to converge more effectively?
16. What activation function is commonly used in the output layer for multiclass classification?
18. What technique involves randomly dropping out some neurons during training to prevent overfitting?