Learn Before
Short Answer

State the significance of the unseen training-distribution error.

Question: In the data mismatch example, the algorithm achieves a 1.5% error rate on unseen data from the training-set distribution. What does this specific metric tell you about the algorithm's performance?

Sample answer: It tells you that the algorithm generalizes well to the training-set distribution, meaning it does not suffer from a severe high variance (overfitting) problem on that specific distribution.

Key points:

  • The algorithm successfully generalizes to the training data distribution.
  • High variance (overfitting) on the training distribution is ruled out as the primary issue.

Rubric: The answer should mention that the algorithm generalizes well to the training distribution or that variance (overfitting) is not the main issue on that distribution.

0

1

Updated 2026-05-27

Contributors are:

Who are from:

Tags

Machine Learning

Deep Learning

Supervised Learning

Dive into Deep Learning @ D2L

Data Science

Machine Learning Strategy

Machine Learning Yearning @ DeepLearning.AI