Learn Before
Short Answer

How can error analysis suggest new input features?

Question: Answer in one to three sentences.

Sample answer: A careful review of mistakes can show that the model struggles with one specific kind of case. That clue can motivate adding a new input feature designed to help the model distinguish that case from others and reduce that error pattern.

Key points:

  • Inspect errors to find a recurring pattern
  • Use that pattern to decide on a new feature
  • Aim the feature at the troublesome subset of examples

Rubric: The answer should connect error analysis to creating an additional feature that targets a specific recurring category of mistakes.

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