Free AWS MLA-C01 Actual Exam Questions - Question 7 Discussion
A company is building a web-based AI application by using Amazon SageMaker. The application will provide the following capabilities and features: ML experimentation, training, a central model registry, model deployment, and model monitoring. The application must ensure secure and isolated use of training data during the ML lifecycle. The training data is stored in Amazon S3. The company needs to run an on-demand workflow to monitor bias drift for models that are deployed to real-time endpoints from the application. Which action will meet this requirement?
A/B? A sounds solid since SageMaker Clarify is built for bias detection and can run on demand, which matches the requirement well. But B might still work if you want more control by pulling the model-monitor-analyzer image and running custom checks via Lambda. The question asks for on-demand bias drift monitoring, so both have merits, but Clarify feels more straightforward for this use case.
Maybe B since pulling the built-in analyzer image lets you customize bias checks flexibly.
A imo. SageMaker Clarify is designed specifically for bias detection and supports running on-demand jobs easily, which fits the question’s need for an on-demand workflow. Lambda triggering a Clarify job keeps it neat and automated without extra manual steps. B might work technically, but it’s more complex and not as straightforward as using Clarify’s built-in capabilities.
Option A seems right since SageMaker Clarify is built exactly for bias detection and supports on-demand jobs, unlike the model-monitor-analyzer which is more for automated monitoring setups.
D imo, using notebooks for direct comparison gives more control and flexibility for on-demand bias drift checks without extra overhead or relying on separate Lambda functions.
B. The model-monitor-analyzer image is built specifically for analyzing model data for drift and bias, making it a good fit for on-demand monitoring without needing the full Clarify job setup.
A makes the most sense here because SageMaker Clarify is specifically designed for bias detection and can be easily triggered on-demand through Lambda, fitting the workflow. B mentions a built-in model monitor image, but that’s more for data quality and drift detection rather than detailed bias analysis. C and D don’t really fit since Glue Data Quality is not tailored to bias monitoring, and notebooks are manual and less scalable for this use case. Overall, A is the cleanest way to integrate on-demand bias drift checks within the app’s architecture.
It’s A, since Lambda triggering SageMaker Clarify fits on-demand bias drift checks best.
A imo, SageMaker Clarify is built for bias checks and works well on-demand via Lambda.
D imo, SageMaker notebooks could be a good manual option for bias comparison without adding extra Lambda complexity. But it's definitely more hands-on and less automated than A or B.
A makes the most sense here because SageMaker Clarify is specifically built for bias detection and works well with Lambda for on-demand jobs. The other options don’t offer as direct or integrated a solution.
Maybe B could fit here too since the sagemaker-model-monitor-analyzer image is built specifically for analyzing model data and detecting bias or drift. It might give a more integrated solution within SageMaker’s model monitoring framework compared to just running Clarify jobs. Plus, using that built-in image via Lambda can automate the bias checks on demand without needing manual notebook work, which rules out D. AWS Glue Data Quality (C) doesn’t really focus on model bias specifically, so probably not the best for this use case.
Maybe B could work too since it involves using the sagemaker-model-monitor-analyzer image, which is designed for monitoring model quality and bias. It might be a more specialized way to run bias drift checks compared to Lambda just triggering Clarify jobs. A sounds good, but B seems like a direct way to handle model monitoring without needing an extra Clarify job setup. Might depend on how integrated you want the monitoring workflow to be with SageMaker’s built-in monitoring features.
Sounds like A fits best for on-demand bias monitoring with SageMaker Clarify. A