AWS Certified Machine Learning Engineer Associate · Free Practice Question Medium
Question 42
Which of the following types of drift is best described as a systematic favoring of certain groups in the model's predictions over time?
-
A
Model quality drift
-
B
Feature attribution drift
-
C
Bias drift
-
D
Data drift
Reveal correct answer
Correct answer: C
A.
Model quality drift refers to a decline in the model’s performance, such as accuracy, and is not specifically related to bias.
B.
Feature attribution drift involves changes in the influence of input features on the model’s predictions, not systematic bias.
C.
Bias drift occurs when a model begins to systematically favor specific groups or categories in its predictions. This can lead to unfair outcomes and reduce the ethical performance of the model.
D.
Data drift refers to changes in the input data, not systematic favoritism in predictions.
Discussion
Think the marked answer is wrong, or have a better explanation? Share it below — comments appear after review.
