Free AWS AIP-C01 Actual Exam Questions - Question 4 Discussion
Scenario: A data scientist needs to develop a fraud detection model on SageMaker with a severely imbalanced dataset (fraudulent transactions are rare). They must minimize operational overhead and ensure the model is fair and unbiased. Question- Which approach will fulfill the given requirements?. Options:
D—SageMaker Pipelines cut overhead, and Clarify fits bias detection needs well.
D Using SageMaker Pipelines reduces manual steps and Clarify is built for bias detection, which fits the need to minimize overhead and ensure fairness better than options involving A2I’s human reviews.
D/C? D’s pipeline automation is a plus, but C also covers SMOTE and bias checks with A2I, which might add a layer of human oversight if needed. Not sure if human review is mandatory though.
Maybe B makes more sense since Amazon A2I can involve human reviewers to catch bias, which fits fairness better, and it keeps overhead lower than building custom pipelines.
D makes sense because it uses SageMaker Pipelines to automate the workflow and Clarify for bias detection, which fits the requirement to minimize operational overhead and check fairness properly.
B tbh, Amazon A2I is mainly for human review workflows, not really bias detection. That makes B a bit off since bias checks here need something like SageMaker Clarify. Also, SMOTE integration in Studio is fine, but operational overhead might be higher if you have to set up A2I manually. D looks cleaner with Pipelines automating the flow and Clarify built in for bias checks, minimizing manual steps.
Option D avoids extra manual steps by using SageMaker Pipelines and Clarify directly.
D seems the best since DeepAR can learn shared patterns across SKUs and handle cold-start forecasting. The others don’t really fit the cold-start scenario well.