#incompleteness

1 question from Explainable AI

Timed test

1 question · 1 min · score and explanations at the end

Q 1 / 1Score 0 / 0

Shuffle questions

Question 1: Single choice

mediumAI-checkedExplainable AI

Doshi-Velez and Kim argue that not every machine learning system needs to explain itself. In which situation is an explanation most clearly needed?

When goals like fairness or safety cannot be fully put into the loss1
When wrong outputs have no significant consequences, as for an ad server2
When the uncertainty of each prediction can be quantified precisely3
When the system has been validated extensively in real use for years4

Keys: 1–9 select · Enter check / next