Regularization

In this video, we learn about regularization: an automatic way to adjust the size of the hypothesis set.

Try answering these comprehension questions to further grill in the concepts covered in this video:

1. What are the reasons to use regularization instead of just manually changing the hypothesis set?
2. Why does regularization hurt training error?
3. Does regularization always help generalization error?
4. How do you find the sweet spot for regularization?
5. In what way is regularization like choosing a hypothesis set?