A company has an ML model that uses historical transaction data to predict customer behavior. An ML engineer is optimizing the model in Amazon SageMaker to enhance the model's predictive accuracy…

Question

Options

• AUse Amazon CloudWatch to monitor network metrics and CPU metrics for resource optimization
• BCreate AWS Glue DataBrew recipes to correct the data based on statistics from the model output.
• CUse SageMaker Clarify to evaluate the model and training data for underlying patterns that might
• DCreate AWS Lambda functions to automate data pre-processing and to ensure consistent quality

• A
  15% (5)
• B
  6% (2)
• C
  76% (26)
• D
  3% (1)

Explanation

SageMaker Clarify is purpose-built for exactly this use case - it analyzes both training data and model predictions to detect bias and identify statistical patterns across demographic groups, directly addressing the requirement to examine inputs and outputs for trends that could skew performance across demographics.

Why the distractors fail:

• A (CloudWatch) monitors infrastructure metrics like CPU and network - useful for operational health, but blind to data patterns or model fairness.
• B (Glue DataBrew) can clean and transform data, but it operates on the data itself, not on model outputs, and doesn't analyze predictions for demographic skew.
• D (Lambda) can automate pre-processing pipelines but has no capability to evaluate model behavior or detect bias patterns in predictions.

Memory tip: Think of Clarify as a bias magnifying glass - it sits between your data and your predictions and asks "is this model treating different groups fairly?" Whenever an exam question mentions demographics, bias, fairness, or examining both inputs and outputs, Clarify is almost certainly the answer.

No community discussion yet for this question.