Practice Transfer Learning (8.6) - Deep Learning and Neural Networks - Data Science Advance

Practice: Transfer Learning

Lesson 8.6Deep Learning and Neural Networks

Practise this section

9 questions

Flashcard drill
3 cards from this section. Quick, good the night before a test.
Exercise
6 questions · about 4 min · pass mark 70%. Hints available.
Sign up to start the exercise
Quiz
3 questions · about 2 min · pass mark 70%. Hints available.
Sign up to start the quiz

6 sample questions

Question 1.
What is transfer learning?
Show hintHide hint
Think about how existing knowledge can help in new situations.
Question 2.
Name one benefit of transfer learning.
Show hintHide hint
What advantage does using existing models provide?
3Question 3.
What is the main goal of transfer learning?
- To create models from scratch
- To reuse knowledge from pre-trained models
- To reduce model complexity
Show hintHide hint
Think about the process of starting a new task; how can pre-existing skills help?
4Question 4.
True or False: Fine-tuning often involves training all layers of a pre-trained model.
- True
- False
Show hintHide hint
Consider which parts of a model carry learned knowledge.
Question 5.
Design a step-by-step plan to use transfer learning on a small dataset for a fruit classification task, using a pre-trained ResNet model.
Show hintHide hint
Consider what parts of the model are important to adapt versus those to keep unchanged.
Question 6.
Evaluate how transfer learning can impact the training of a chatbot using a limited conversation dataset by leveraging a pre-trained BERT model.
Show hintHide hint
Think about why BERT's extensive training makes it a suitable choice for quick adaptation.

5 more questions in this section. Enroll free to practise them all.