Short Answer

Meaning of error on the training-like distribution

Question: In a credit-scoring example, the model achieves a 2.3% error rate on a holdout set drawn from the same distribution as the training data. What does this metric indicate about the model?

Sample answer: It indicates that the model generalizes well to that distribution, so severe overfitting is not the main problem on that data source.

Key points:

  • The model performs well on the training-like distribution.
  • High variance is not the primary issue on that distribution.

Rubric: The answer should state that the model does well on the training-like distribution or that overfitting/high variance is not the main issue there.

0

1

Updated 2026-08-12

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