Short Answer

Why is manual review of mistakes less useful on tasks people cannot do reliably?

Question: Why does a human review of model errors become less helpful when the underlying task is one that people cannot perform well themselves?

Sample answer: It is harder to understand why the model made a mistake when humans do not have strong skill at the task. If people cannot reliably do the task, they are less able to judge the cause of each error.

Key points:

  • Harder to identify the cause of a wrong prediction.
  • The task is beyond human ability or human performance is poor.

Rubric: The response must state that it is difficult to determine why the model made an error when humans cannot perform the task well.

0

1

Updated 2026-08-12

Contributors are:

Who are from:

Tags

Machine Learning

Deep Learning

Machine Learning Strategy

Supervised Learning

Dive into Deep Learning @ D2L

Data Science

Machine Learning Yearning @ DeepLearning.AI