AWS Certified Machine Learning Engineer Associate · Free Practice Question Medium
Question 45
What is the key challenge that results from a high variance model in the context of responsible AI, and what technique is most effective to combat it?
-
A
Underfitting, and the solution is cross-validation
-
B
Overfitting, and the solution is early stopping
-
C
Overfitting, and the solution is adding more training data
-
D
Underfitting, and the solution is reducing the complexity of the model
Reveal correct answer
Correct answer: B
A.
Cross-validation is useful for generalization but does not specifically address high variance leading to overfitting.
B.
A high variance model results in overfitting, meaning the model performs very well on training data but fails to generalize to new data. Early stopping is an effective technique to prevent the model from continuing to overfit during training.
C.
Adding more training data can help with underfitting, but the main concern here is overfitting due to high variance.
D.
Reducing model complexity may help with overfitting, but the correct technique for high variance is early stopping, which directly tackles the issue during training.
Discussion
Think the marked answer is wrong, or have a better explanation? Share it below — comments appear after review.
