MLA-C01 Question Bank · Latest 50 Questions Free Preview (Page 5/5)
Questions 41–50 of the latest 50 MLA-C01 questions (5 pages total), with answers and explanations, no login required. Practice all 233 questions online.
Question 41 · No. 224 · Single choice
A hospital is using an ML model to validate x-ray results. The hospital runs a nightly batch inference job. The hospital needs to produce a daily report about model data quality and model performance. Which solution will meet these requirements?
- A. Schedule a monitoring job in Amazon SageMaker Model Monitor. Generate the monitoring results for the model and data.
- B. Create an Amazon CloudWatch dashboard that includes the metrics for processing steps in the nightly batch inference job. Compare the baseline resource metrics. Share the dashboard link.
- C. Use AWS Glue DataBrew to create a custom recipe job that uses the Numerical Statistics data quality check for the model file. Generate the results.
- D. Create a SageMaker AI pipeline that includes a QualityCheck step to run monitoring jobs. Generate the monitoring results for the model and the data.
Answer: D
Question 42 · No. 225 · Single choice
A company runs an Amazon SageMaker AI domain in a public subnet of a newly created VPC. The network is configured properly, and ML engineers can access the SageMaker AI domain. Recently, the company discovered suspicious traffic to the domain from a specific IP address. The company needs to block traffic from the specific IP address. Which update to the network configuration will meet this requirement?
- A. Create a security group inbound rule to deny traffic from the specific IP address. Assign the security group to the domain.
- B. Create a network ACL inbound rule to deny traffic from the specific IP address. Assign the rule to the default network ACL for the subnet where the domain is located.
- C. Create a shadow variant for the domain. Configure SageMaker Inference Recommender to send traffic from the specific IP address to the shadow endpoint.
- D. Create a VPC route table to deny inbound traffic from the specific IP address. Assign the route table to the domain.
Answer: B
Explanation: Network ACLs (NACLs) are stateless and support both allow and deny rules, making them suitable for blocking traffic from a specific IP address at the subnet level. Security groups (option A) only support allow rules and cannot explicitly deny traffic from specific IPs. VPC route tables (option D) cannot deny traffic from specific IP addresses as they are used for directing traffic destinations, not filtering source IPs. Shadow variants (option C) are used for A/B testing and model deployment, not for blocking traffic. Therefore, creating a network ACL inbound deny rule is the correct approach.
Question 43 · No. 226 · Single choice
A company is using Amazon SageMaker AI to develop a credit risk assessment model. During model validation, the company finds that the model achieves 82% accuracy on the validation data. However, the model achieved 99% accuracy on the training data. The company needs to address the model accuracy issue before deployment. Which solution will meet this requirement?
- A. Add more dense layers to increase model complexity. Implement batch normalization. Use early stopping during training.
- B. Implement dropout layers. Use L1 or L2 regularization. Perform k-fold cross-validation.
- C. Use principal component analysis (PCA) to reduce the feature dimensionality. Decrease model layers. Implement cross-entropy loss functions.
- D. Augment the training dataset. Remove duplicate records from the training dataset. Implement stratified sampling.
Answer: B
Explanation: The model achieves 99% accuracy on training data but only 82% on validation data, indicating classic overfitting. Option B directly addresses overfitting through three complementary techniques: dropout layers randomly deactivate neurons during training to prevent the model from becoming overly dependent on specific features; L1/L2 regularization adds penalty terms to the loss function to constrain model weights and reduce complexity; and k-fold cross-validation provides robust evaluation to ensure the model generalizes well. Options A, C, and D either worsen overfitting (adding complexity) or include less relevant techniques. Option B is the most comprehensive and appropriate solution.
Question 44 · No. 227 · Single choice
An ML engineer wants to use, prepare, and load data from Amazon S3 for analytics. The ML engineer must run an extract, transform, and load (ETL) job to discover the schema of the data and to store the metadata. Which solution will meet these requirements with the LEAST manual effort?
- A. Use AWS Glue to run the ETL job. Use the job to discover the schema and to store the associated metadata in the AWS Glue Data Catalog.
- B. Create an Amazon SageMaker Data Wrangler flow to run the ETL job. Use the job to discover the schema and to store the associated metadata in an S3 bucket.
- C. Create an ETL pipeline by using Amazon Athena integrated with AWs Step Functions. Use the pipeline to run the ETL job to discover the schema and to store the associated metadata in an S3 bucket.
- D. Launch an Amazon EC2 instance that includes the scikit-learn library to run the ETL job. Use the job to discover the schema and to store the associated metadata in Amazon Redshift.
Answer: A
Explanation: AWS Glue is a fully managed extract, transform, and load (ETL) service designed to automatically discover the schema of data in Amazon S3 and store the associated metadata in the AWS Glue Data Catalog. This serverless approach requires the least manual effort compared to the other options, which either involve more manual configuration (SageMaker Data Wrangler), orchestrating multiple services (Athena with Step Functions), or managing infrastructure yourself (EC2 with scikit-learn). Therefore, option A is the most efficient solution for the given requirements.
Question 45 · No. 228 · Single choice
A logistics company has installed in-vehicle cameras for basic monitoring of its drivers. The company wants to improve driver safety by identifying distractions that could lead to accidents. Which solution will meet this requirement with the LEAST operational effort?
- A. Use Amazon Rekognition eye gaze direction detection to monitor driver behavior and identify distractions.
- B. Use Amazon SageMaker AI to customize an AI model to monitor driver behavior and identify distractions.
- C. Integrate a third-party driver monitoring system with Amazon Rekognition to monitor driver behavior and identify distractions
- D. Use Amazon Comprehend to analyze text-based driver feedback and identify distractions.
Answer: A
Explanation: Amazon Rekognition is a fully managed AI service that provides pre-built capabilities, including eye gaze direction detection. This ready-to-use feature requires minimal operational effort as it eliminates the need to build, train, and deploy custom models (unlike Amazon SageMaker). It also avoids the complexity of integrating third-party systems (unlike option C) and is appropriate for analyzing video feeds from in-vehicle cameras (unlike Amazon Comprehend, which is designed for text analysis). Therefore, option A offers the most straightforward solution with the least operational overhead.
Question 46 · No. 229 · Single choice
An ML engineer is analyzing a classification dataset before training a model in Amazon SageMarker AI. The ML engineer suspects that the dataset has a significant imbalance between class labels that could lead to biased model predictions. To confirm class imbalance, the ML engineer needs to select an appropriate pre-training bias metric. Which metric will meet this requirement?
- A. Mean square error (MSE)
- B. Difference in proportions of labels (DPL)
- C. Silhouette score
- D. Structural similarity index measure (SSIM)
Answer: B
Explanation: Difference in Proportions of Labels (DPL) is a pre-training bias metric in Amazon SageMaker Clarify specifically designed to detect imbalance between class labels in a dataset. MSE is for regression, Silhouette score is for clustering evaluation, and SSIM is for image similarity comparison. Since the ML engineer is analyzing class imbalance in a classification dataset and needs a pre-training bias metric, DPL is the correct choice.
Question 47 · No. 230 · Single choice
A company has an ML model that is deployed to an Amazon SageMaker endpoint for real-time inference. The company needs to deploy a new model. The company must compare the new model’s performance to the currently deployed model’s performance before shifting all traffic to the new model. Which solution will meet these requirements with the LEAST operational effort?
- A. Deploy the new model to a separate endpoint. Manually split traffic between the two endpoints.
- B. Deploy the new model to a separate endpoint. Use Amazon CloudFront to distribute traffic between the two endpoints.
- C. Deploy the new model as a shadow variant on the same endpoint as the current model. Route a portion of live traffic to the shadow model for evaluation.
- D. Use AWS Lambda functions with custom logic to route traffic between the current model and the new model.
Answer: C
Explanation: Amazon SageMaker provides built-in support for shadow variants, which allows you to deploy a new model alongside the current production model on the same endpoint. Shadow variants receive a portion of live traffic (or a copy of the inference requests) for evaluation purposes without affecting the responses sent back to clients. This is a managed feature that requires minimal operational effort compared to deploying separate endpoints, manually splitting traffic, setting up CloudFront routing, or writing custom Lambda logic. Option A requires manual traffic management, Option B misuses CloudFront which is designed for content delivery rather than ML inference routing, and Option D requires writing and maintaining custom routing logic.
Question 48 · No. 231 · Single choice
A company is using Amazon SageMaker AI to create a classification model to categorize the company’s sales performance for each month of the previous 20 years on a scale from 1 to 5. The dataset includes fields for month, sales region, regional aggregate sales, and the number of stores in each sales region. The company notices that during two months of every year, the aggregate sales values are unexpectedly high. The company performs one-hot encoding on all non-numerical features in the training and validation datasets. The company uses the training dataset to train the classification model. When the company evaluates the model against the validation dataset, the results are less accurate than expected. The company must improve the model’s accuracy on the validation dataset. Which solution will meet this requirement?
- A. Remove records that include outliers across all features.
- B. Use a stratified split on the month and sales region features.
- C. Perform normalization on the aggregate sales feature.
- D. Perform normalization on the aggregate sales feature for each sales region.
Answer: D
Question 49 · No. 232 · Single choice
An ML engineer is building an ML model in Amazon SageMaker AI. The ML engineer needs to load historical data directly from Amazon S3, Amazon Athena, and Snowflake into SageMaker AI. Which solution will meet this requirement?
- A. Use AWS Glue DataBrew to import the data into SageMaker AI.
- B. Build a pipeline in SageMaker Pipelines to process the data. Use AWS DataSync to load the processed data into SageMaker AI.
- C. Create a feature store in SageMaker Feature Store. Use an Apache Spark connector to Feature Store to access the data.
- D. Use SageMaker Data Wrangler to query and import the data.
Answer: D
Explanation: SageMaker Data Wrangler is the correct solution because it natively supports direct data import from Amazon S3, Amazon Athena, and Snowflake. It provides a visual interface that allows ML engineers to query, transform, and import data directly into SageMaker AI for model building. Option A (Glue DataBrew) is mainly for data preparation, not direct loading into SageMaker AI. Option B (SageMaker Pipelines with DataSync) is inappropriate because DataSync is designed for transferring data between storage services, not for loading data into SageMaker AI from query sources like Athena and Snowflake. Option C (Feature Store with Spark connector) is designed for feature management and retrieval, not for initial data ingestion from multiple sources.
Question 50 · No. 234 · Single choice
An ML model is deployed in production. The model has performed well and has met its metric thresholds for months. An ML engineer who is monitoring the model observes a sudden degradation. The performance metrics of the model are now below the thresholds. What could be the cause of the performance degradation?
- A. Lack of training data
- B. Drift in production data distribution
- C. Compute resource constraints
- D. Model overfitting
Answer: B
Explanation: The most likely cause of sudden performance degradation in a production ML model that has been performing well is drift in the production data distribution. When the statistical properties of the incoming production data change from what the model was trained on (concept drift or data drift), the model's predictions become less accurate, causing performance metrics to drop below established thresholds. Option A (lack of training data) is incorrect because training data availability doesn't affect a deployed model. Option C (compute resource constraints) would cause latency or availability issues, not prediction accuracy degradation. Option D (overfitting) is a training-time issue that would have been detected before deployment, not something that appears suddenly after months of good performance.