Question 1

Which of the following is not an algorithm for training word vectors?

Options :

A : TextCNN

B : BERT

C : FastText

D : Word2Vec

Answer: A

Question 2

Which of the following statements are true about the differences between using convolutional neural networks (CNNs) in text tasks and image tasks?

Options :

A : Color image input is multi-channel, whereas text input is single-channel.

B :

When the CNN is used for text tasks, the kernel size must be the same as the number of word vector dimensions. This constraint, however, does not apply to image tasks.

C : For CNN, there is no difference in handling text or image tasks.

D : CNNs are suitable for image tasks, but they perform poorly in text tasks.

Answer: A,B

Question 3

In the deep neural network (DNN)–hidden Markov model (HMM), the DNN is mainly used for feature processing, while the HMM is mainly used for sequence modeling.

Options :

A : TRUE

B : FALSE

Answer: A

Question 4

John wants to deploy a large model locally to implement the Q&A assistant function for his company. Which of the following factors is unnecessary for John to consider?

Options :

A : Model development framework

B : Output delay

C : Model security

D : Demand for computing power

Answer: A

Question 5

Which of the following statements about the functions of layer normalization and residual connection in the Transformer is true?

Options :

A :

Residual connections and layer normalization help prevent vanishing gradients and exploding gradients in deep networks.

B : Residual connections primarily add depth to the model but do not aid in gradient propagation.

C : Layer normalization accelerates model convergence and does not affect model stability.

D :

In shallow networks, residual connections are beneficial, but they aggravate the vanishing gradient problem in deep networks.

Answer: A

Free H13-321_V2.5 Mock Exam – Practice Online Confidently

Buying Options: