Case Study

Sharing Only the Relevant Pages to Explain a Model Choice

Case context: A project lead recommends collecting more labeled examples before spending time on architecture tuning. A teammate is unsure why that tradeoff makes sense, but only has a few minutes to review documentation.

Question: According to the short-document strategy, what should the lead do so the teammate can understand and judge the recommendation?

Sample answer: The lead should pick the short section that explains why more training data is the better next step, then share only the one or two pages that matter. That gives the teammate enough context to evaluate the recommendation without forcing them to read the entire document.

Key points:

  • Select the part that directly supports the recommendation
  • Share only the one or two relevant pages
  • Keep the reading burden small because the teammate has limited time
  • Make it possible for the teammate to understand and evaluate the decision

Rubric: Strong answers explain that only the relevant one or two pages should be shared, connect that choice to the teammate's time limit and uncertainty, and show how it helps the teammate assess the recommendation.

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