Amazon Web Services MLA-C01 - AWS Certified Machine Learning Engineer - Associate

One-hot encoding is the appropriate technique for transforming categorical data, such as color information, into a format suitable for input to a neural network. This technique creates a binary vector representation where each unique category (color) is represented as a separate binary column, ensuring that the model does not infer ordinal relationships between categories. This approach preserves the categorical nature of the data and avoids introducing unintended biases.

The key requirements are real-time processing, high throughput, and minimal operational overhead. Amazon Kinesis Data Streams is designed for ingesting thousands of events per second with low latency.

For anomaly detection on streaming data, Amazon Managed Service for Apache Flink provides a built-in Random Cut Forest (RCF) function. RCF is an unsupervised anomaly detection algorithm that works well on numerical streaming data and does not require labeled training data.

This fully managed combination eliminates the need to deploy or maintain SageMaker endpoints, EC2 instances, or custom ML pipelines. Options B and C introduce unnecessary infrastructure and model management overhead. Option D is batch-oriented and unsuitable for real-time anomaly detection.

Therefore, using Kinesis Data Streams with Flink’s built-in Random Cut Forest is the most scalable and low-overhead solution.

When running consecutive training jobs in Amazon SageMaker, infrastructure provisioning can introduce latency, as each job typically requires the allocation and setup of compute resources. To minimize this startup time and enhance efficiency, Amazon SageMaker offers Managed Warm Pools.

Key Features of Managed Warm Pools:

Reduced Latency: Reusing existing infrastructure significantly reduces startup time for training jobs.

Configurable Retention Period: Allows retention of resources after training jobs complete, defined by the KeepAlivePeriodInSeconds parameter.

Automatic Matching: Subsequent jobs with matching configurations (e.g., instance type) can reuse retained infrastructure.

Implementation Steps:

Request Warm Pool Quota Increase: Increase the default resource quota for warm pools through AWS Service Quotas.

Configure Training Jobs:

Set KeepAlivePeriodInSeconds for the first training job to retain resources.

Ensure subsequent jobs match the retained pool ' s configuration to enable reuse.

Monitor Warm Pool Usage: Track warm pool status through the SageMaker console or API to confirm resource reuse.

Considerations:

Billing: Resources in warm pools are billable during the retention period.

Matching Requirements: Jobs must have consistent configurations to use warm pools effectively.

Alternative Options:

Managed Spot Training: Reduces costs by using spare capacity but doesn’t address startup latency.

SageMaker Training Compiler: Optimizes training time but not infrastructure setup.

SageMaker Distributed Data Parallelism Library: Enhances training efficiency but doesn’t reduce setup time.

By using Managed Warm Pools, the company can significantly reduce startup latency for consecutive training jobs, ensuring faster experimentation cycles with minimal operational overhead.

AWS Documentation: Managed Warm Pools

AWS Blog: Reduce ML Model Training Job Startup Time

Amazon SageMaker Serverless Inference is designed for irregular or unpredictable traffic patterns. It automatically provisions and scales compute resources based on request volume and scales down to zero when idle, making it the most cost-effective option.

Serverless inference supports payloads up to 6 MB and request durations up to 60 seconds, which comfortably meets the stated constraints. Customers are billed only for actual compute usage during inference execution, not for idle capacity.

Asynchronous inference is intended for long-running jobs (up to 1 hour) and large payloads (up to 1 GB). Real-time inference requires always-on instances, increasing cost during idle periods. Batch transform is designed for offline processing.

Therefore, serverless inference is the optimal choice.

AWS Compute Optimizer analyzes the resource usage of Amazon EC2 instances, ECS services, Lambda functions, and Amazon EBS volumes. It provides actionable recommendations to optimize resource utilization and reduce costs, such as resizing instances, moving workloads to Spot Instances, or changing volume types. This solution requires the least development effort because Compute Optimizer is a managed service that automatically generates insights and recommendations based on historical usage data.

The correct answer is B. Use the Amazon SageMaker Model Registry to catalog the models. Create model groups for each model to manage the model versions and to maintain associated metadata.

The Amazon SageMaker Model Registry is a managed repository within SageMaker designed specifically for production-grade ML model lifecycle management. It allows organizations to catalog models, track multiple versions of a model, associate rich metadata, and manage deployment workflows in a scalable, controlled manner. Each model can belong to a model group, which acts as a container for all versions of that particular model. Versions can store training metrics, hyperparameters, model artifacts, and other key metadata, enabling reproducibility, auditing, and automated promotion between stages (e.g., Staging → Production).

Option A, while using the Model Registry, relies on manually tagging versions and creating key-value pairs to store metadata. This approach is error-prone, lacks structured versioning, and does not integrate with SageMaker’s deployment pipelines.

Options C and D suggest using Amazon ECR repositories. While ECR can store containerized model artifacts, it is not designed for ML-specific metadata, versioning, or automated model stage transitions. Using ECR alone would require custom-built solutions for metadata management, auditing, and version tracking, adding unnecessary operational overhead.

By leveraging the Model Registry with model groups, organizations can automate promotions, apply approval workflows, and track lineage efficiently, fully aligning with AWS best practices for ML model development and production readiness. This ensures compliance, reproducibility, and reduces operational complexity in enterprise AI platforms.

Using the Model Registry and model groups is the standard AWS-recommended approach for enterprise-scale model cataloging and version control, enabling teams to focus on model improvement rather than infrastructure management.

The primary goal is to produce the most complete dataset while handling missing values and extreme outliers appropriately. Dropping samples (Option A) or columns (Option D) would reduce data completeness and potentially remove valuable information, which contradicts the requirement.

Imputation is therefore the correct approach. Between mean and median imputation, AWS ML best practices recommend using the median when features contain outliers. The mean is sensitive to extreme values and can be skewed significantly, leading to imputed values that are not representative of the typical data distribution. In contrast, the median is robust to outliers, making it a better statistical estimator for central tendency in such datasets.

Amazon SageMaker Data Wrangler supports median imputation as a built-in transformation, enabling ML engineers to handle missing values consistently across large tabular datasets without custom code. This approach preserves all rows and columns while minimizing distortion caused by extreme values, which is particularly important for neural networks that are sensitive to input distributions.

Therefore, imputing missing values with the median value provides the most complete and statistically appropriate dataset for training.

AWS Glue DataBrew is specifically designed to provide no-code and low-code data preparation for analytics and machine learning. It supports common file formats such as Apache Parquet and integrates directly with Amazon S3.

Using DataBrew, users can visually create recipes that apply transformations such as one-hot encoding without writing any code. Once the recipe is defined, a DataBrew job can be run to process the dataset and store the transformed output back into Amazon S3.

Options B, C, and D all require writing SQL or code, which violates the no-code requirement. AWS documentation clearly identifies DataBrew as the correct service for interactive, visual data transformation at scale.

Therefore, Option A is the correct solution.

Hyperband is a hyperparameter tuning strategy designed to minimize computation time by adaptively allocating resources to promising configurations and terminating underperforming ones early. It efficiently balances exploration and exploitation, making it ideal for large datasets and deep learning models where training can be computationally expensive.

Large Parquet files can cause out-of-memory (OOM) issues during training if individual files exceed the memory capacity of the training instances. AWS documentation recommends repartitioning large datasets into smaller files to enable efficient streaming and parallel loading.

By using Apache Spark on Amazon EMR to repartition the Parquet files, the ML engineer can split the dataset into multiple smaller files that can be read incrementally during training. This approach avoids loading a single large file into memory and improves data parallelism.

Attaching larger EBS volumes increases storage capacity but does not solve memory constraints. Switching to memory-optimized instances increases cost and is not necessary when the dataset can be restructured. SageMaker distributed data parallelism focuses on model parameter synchronization across GPUs, not dataset file size.

AWS best practices explicitly recommend partitioning large Parquet datasets to improve memory efficiency during training.

Therefore, Option B is the correct and AWS-aligned solution.