Essay

Evaluating Unsolvable Errors

Question: Explain why it is beneficial to include error categories in your analysis even if you currently lack a clear idea of how to improve or fix them.

Sample answer: Including error categories that you don't currently know how to improve is beneficial because the primary goal of the error analysis process is to build intuition. By tracking these errors, you can understand their frequency and impact on your model. This information helps you identify the most promising areas to focus your future efforts on, even if you don't have an immediate solution at hand.

Key points:

  • Builds intuition
  • Identifies the most promising areas to focus on
  • Error categories do not require immediate solutions to be tracked

Rubric: A strong answer should explain that tracking these errors helps build intuition and identifies promising areas for future focus, referencing that immediate solutions are not strictly required for a category to be useful.

0

1

Updated 2026-06-18

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