Practice Training Deep Networks (8.3) - Deep Learning and Neural Networks

Practice: Training Deep Networks

Lesson 8.3Deep Learning and Neural Networks

Practise this section

10 questions

Flashcard drill
4 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
4 questions · about 3 min · pass mark 70%. Hints available.
Sign up to start the quiz

6 sample questions

Question 1.
What is backpropagation?
Show hintHide hint
Think about how we correct mistakes after getting our predictions.
Question 2.
What does gradient descent aim to minimize?
Show hintHide hint
Consider what error we want to reduce in our predictions.
3Question 3.
What does backpropagation help optimize in neural networks?
- Weight loss
- Gradient descent
- Loss function
Show hintHide hint
Consider what we are trying to minimize by optimizing our predictions.
4Question 4.
True or False: Overfitting results in better model accuracy on unseen data.
- True
- False
Show hintHide hint
Think about what happens when a model memorizes the training data.
Question 5.
You have a neural network with 5 hidden layers. During training, it experiences vanishing gradients, which seem to halt learning. Describe methods you might implement to alleviate this issue.
Show hintHide hint
Consider how architectures help control gradient behavior.
Question 6.
Design an experiment comparing the effectiveness of Batch Gradient Descent to Mini-batch Gradient Descent on a set of training data. Outline the criteria you would measure and hypothesize expected outcomes.
Show hintHide hint
Think about the practical implications of each method's data processing.

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