Which Amazon SageMaker feature provides a centralized repository for storing, sharing, and discovering ML features for both training and real-time inference?
Aws Ml Engineer Associate Sagemaker Platform
Question · Medium
An ML engineer needs to detect statistical bias in training data and explain individual predictions using SHAP values. Which SageMaker capability supports both requirements?
Aws Ml Engineer Associate Responsible Ai
Chapter 1: Introduction & Exam Strategy
1.1 AWS MLA-C01 Blueprint, Format & Strategy
Study guide
Which SageMaker inference option is most cost-effective for sporadic, unpredictable traffic where cold-start latency is acceptable?
Aws Ml Engineer Associate Deployment Orchestration
Tech Exam Prep Podcast
Listen on Spotify
Podcast
A team must process inference requests up to 1 GB in size with processing times of up to 60 minutes per request. Which SageMaker endpoint type is appropriate?
Question · Hard
Which AWS service provides a visual data preparation tool that lets analysts clean and normalize datasets with 250+ pre-built transformations and no code?
Aws Ml Engineer Associate Data Preparation
Question · Easy
An ML engineer wants to run hyperparameter tuning that adapts intelligently based on past trial results to minimize total trials needed. Which SageMaker tuning strategy should they choose?
Aws Ml Engineer Associate Model Development
1.2 The Machine Learning Engineering Lifecycle on AWS
Which SageMaker tuning strategy uses early-stopping of underperforming trials to allocate budget to promising configurations and is best for deep learning?
Which Amazon Bedrock feature lets you connect a foundation model to enterprise data sources for retrieval-augmented generation (RAG) without building your own ingestion pipeline?
Aws Ml Engineer Associate Generative Ai
A binary classifier predicts fraud (positive class) on a highly imbalanced dataset. The business cares about catching as many fraudulent transactions as possible, even at the cost of false alarms. Which metric should be optimized?
Which SageMaker capability tracks lineage across data, code, hyperparameters, and metrics for ML training runs to make experiments reproducible and comparable?
Aws Ml Engineer Associate Mlops
Domain 1: Data Ingestion & Storage Architecture for ML
2.1 Ingesting Batch and Streaming Data for ML
An organization wants to deploy hundreds of small per-tenant models behind a single endpoint to minimize cost. Which SageMaker hosting option is best?
Which SageMaker Model Monitor type detects shifts in the statistical distribution of inference inputs compared to a training baseline?
Aws Ml Engineer Associate Monitoring Maintenance
A training job on a single ml.p4d.24xlarge instance is bottlenecked by GPU memory because the model is too large to fit. Which distributed training strategy is appropriate?
Which AWS service is designed to label training data using a combination of human workforces (private, vendor, or Mechanical Turk) and active learning to reduce labeling costs?
2.2 High-Performance Storage for Training and Inference
A regression model predicts continuous house prices. Which metric most heavily penalizes large errors?
Which SageMaker capability provides pre-built templates and access to hundreds of foundation models, including Llama, Mistral, and Stable Diffusion, for one-click deployment or fine-tuning?
Which AWS service catalog stores metadata about datasets in a central repository and is automatically updated by Glue crawlers?
Which SageMaker capability orchestrates ML workflows as DAGs of processing, training, evaluation, and registration steps, integrated with the Model Registry?
2.3 Data Cataloging, Governance, and Querying
What technique uses synthetic samples for the minority class generated from k-nearest-neighbor interpolation to address class imbalance?
Which Amazon Bedrock capability lets administrators block harmful content categories, filter sensitive PII, and prevent prompt injection attacks against deployed apps?
Which encoding technique converts a categorical feature with three values (red, green, blue) into three binary columns?
Which SageMaker service surfaces real-time training-time issues such as vanishing/exploding gradients, overfitting, and tensor anomalies via built-in rules?
Domain 1: Feature Engineering & Data Transformation
3.1 Distributed Feature Engineering with AWS Glue & DataBrew
What is the purpose of L2 regularization in a linear or neural model?
An ML engineer wants to query data sitting in S3 using standard SQL without managing servers. Which AWS service is designed for this?
Which SageMaker capability is a no-code visual interface that lets business analysts build ML models without writing Python or SQL?
Which SageMaker AutoML capability automatically tries multiple algorithms, preprocesses data, and produces a leaderboard of candidate models from a tabular dataset?
3.2 Interactive Data Preparation with SageMaker Data Wrangler
Which SageMaker security feature isolates training and inference network traffic to a customer VPC with no internet egress?
Aws Ml Engineer Associate Security
Which SageMaker training mode lets engineers bring their own Docker container with custom dependencies but still benefits from managed training infrastructure?
Which validation strategy partitions data into k folds and trains the model k times to estimate generalization error more reliably than a single holdout?
Which SageMaker feature lets you progressively shift inference traffic from an old model to a new model and roll back automatically on CloudWatch alarm breaches?
3.3 Feature Management with SageMaker Feature Store
Which dimensionality-reduction technique projects high-dimensional data onto axes of maximum variance using eigendecomposition of the covariance matrix?
An ML engineer wants the lowest possible inference latency for a model deployed to edge devices in factories with intermittent connectivity. Which AWS solution is appropriate?
Which Bedrock model customization technique trains additional layers/weights on a labeled task-specific dataset to improve accuracy on a narrow domain?
Which CloudWatch capability allows automatic scaling of a SageMaker endpoint based on InvocationsPerInstance metric thresholds?
Domain 1: Data Quality, Integrity & Ground Truth Labeling
4.1 Data Validation, Schema Drift, and Quality Rules
Which optimizer adapts the learning rate per parameter using running averages of past gradients and squared gradients?
What is the best SageMaker option for one-time scoring of a 5 TB dataset stored in S3 where latency is not important?
Which AWS service provides a serverless event bus that can trigger SageMaker pipelines based on schedules or events such as new objects landing in S3?
Which encryption configuration ensures SageMaker training job artifacts in S3 are encrypted using a customer-managed KMS key?
Loading more…