Short Answer

When are informal bias and variance judgments enough?

Question: Answer in one to three sentences. Make a clear distinction between choosing an action in practice and establishing a mathematical identity.

Sample answer: For making practical decisions on a machine learning project, informal ideas about bias and variance are usually sufficient. Formal definitions are mainly needed when proving the exact error decomposition, and that proof is typically written for mean squared error.

Key points:

  • Informal reasoning can guide practical choices.
  • Formal notation is used for proofs.
  • The exact decomposition is tied to mean squared error.

Rubric: The answer must say that informal bias-variance reasoning is enough for practical decisions and that the formal proof uses mean squared error.

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