4. Neural Methods

Cnns For Nlp — Quiz

Test your understanding of cnns for nlp with 5 practice questions.

Practice Questions

Question 1

When applying a convolutional filter of size $k \times d$ (where $d$ is the embedding dimension) to a sentence, what is the primary characteristic of the feature it extracts?

Question 2

In a CNN for sentence classification, if we use multiple filters with different window sizes (e.g., 2, 3, and 4 words), how does this contribute to the model's ability to understand the sentence?

Question 3

After applying convolutional layers, a pooling layer is typically used. Which of the following best describes the primary goal of this pooling operation in CNNs for NLP?

Question 4

Consider a sentence with $N$ words, each represented by an embedding of dimension $D$. If a convolutional filter has a height of $H$ words and covers the entire embedding dimension $D$, what is the shape of this filter?

Question 5

Which of the following architectural variants of CNNs for NLP is characterized by applying convolutional filters at different 'rates' or 'dilations' to expand the receptive field without increasing the number of parameters?