AIF-C01 Fundamentals of AI and ML Practice Test 2 — 15 Questions

Question 1

Refer to the exhibit. A SageMaker training job fails with an 'AccessDenied' error when trying to read files from the S3 bucket 'my-training-data'. The IAM role used by the training job has the policy shown. What is the most likely reason for the failure?

Choose an answer to begin — your selection is scored in the full session.

Accepted Answer

The policy does not include the s3:ListBucket action. The IAM policy grants s3:GetObject but not s3:ListBucket. When a SageMaker training job reads files from S3, the SageMaker SDK or framework (e.g., TensorFlow, PyTorch) often performs a ListBucket call first to enumerate objects in the prefix. Without s3:ListBucket, the SDK cannot discover the files, resulting in an AccessDenied error even though GetObject is allowed.

Answer

The bucket policy requires encryption in transit

Answer

The training job is using the wrong AWS region

Answer

The policy does not include the s3:PutObject action

Question 2

Refer to the exhibit. A SageMaker real-time endpoint is experiencing increasing latency and memory errors after running for a few hours. What is the most likely cause and recommended fix?

Accepted Answer

Use SageMaker Debugger to identify and fix a memory leak in the inference code. The symptoms—increasing latency and memory errors after running for a few hours—point to a memory leak in the inference code. SageMaker Debugger can monitor system metrics like memory utilization and detect anomalies, helping to identify the root cause of the leak. Fixing the memory leak directly resolves the progressive degradation, whereas scaling or auto-scaling only masks the symptom.

Answer

Scale the endpoint to a larger instance type, such as ml.r5.large

Answer

Enable auto-scaling to add instances during high load

Answer

Use SageMaker Model Monitor to detect data drift

Question 3

A data scientist is building a binary classification model for fraud detection. The dataset is highly imbalanced (99% legitimate, 1% fraud). Which metric is most appropriate to evaluate model performance?

Accepted Answer

F1-score. In highly imbalanced datasets (99% legitimate, 1% fraud), accuracy is misleading because a model that predicts all transactions as legitimate would achieve 99% accuracy but fail to detect any fraud. The F1-score is the harmonic mean of precision and recall, providing a balanced measure that accounts for both false positives and false negatives, making it the most appropriate metric for evaluating fraud detection models.

Answer

Accuracy

Answer

Recall

Answer

Precision

Question 4

A data scientist is using Amazon SageMaker to train a model. The training job is taking longer than expected. Which change would most likely reduce training time?

Accepted Answer

Use a larger batch size. Using a larger batch size allows the model to process more training samples per iteration, which reduces the number of weight updates needed per epoch and can improve hardware utilization (e.g., GPU parallelism). This often leads to faster training times, provided the batch size fits within memory constraints and does not degrade model convergence.

Answer

Increase the number of training epochs

Answer

Use a smaller instance type

Answer

Enable spot training

Question 5

Which TWO of the following are best practices for data preprocessing in machine learning? (Select TWO.)

Accepted Answer

Use cross-validation to evaluate model performance. Option A is correct because cross-validation (e.g., k-fold) provides a more robust estimate of model performance by training and validating on multiple data splits, reducing the variance of the evaluation and helping detect overfitting before final deployment. Option D is correct because gradient-based algorithms such as linear regression with gradient descent, logistic regression, and neural networks converge faster and more reliably when features are on comparable scales, so standardization (z-score) or min-max normalization is a standard preprocessing step. Option B is not a best practice because the 80/20 split is only a common default, not a universal rule; the correct ratio depends on dataset size, and stratified or k-fold schemes are often preferable. Option C is wrong because one-hot encoding is appropriate for nominal categories, whereas ordinal categories should use ordinal/label encoding that preserves their inherent order. Option E is wrong because duplicate records should generally be identified and removed (or investigated) as part of data cleaning regardless of whether they arose from manual entry errors or automated processes, since duplicates bias the model.

Answer

Always split data 80/20 for training and testing

Answer

One-hot encoding for ordinal categories

Answer

Drop duplicate records only if they are manual entry errors

Question 6

Which THREE of the following are capabilities of Amazon SageMaker? (Select THREE.)

Accepted Answer

Real-time inference endpoints. Amazon SageMaker supports real-time inference endpoints (A), which let you deploy a trained model to a persistent HTTPS endpoint for low-latency predictions. It also provides automatic model tuning (B), also called hyperparameter optimization, which runs multiple training jobs to find the best hyperparameter configuration. SageMaker additionally offers built-in algorithms (D) for common tasks such as linear regression, k-means clustering, XGBoost, and image classification, so you can train without writing custom algorithm code. Option C is incorrect because SageMaker is a fully managed cloud service, not an on-premises-only training tool. Option E is incorrect because SageMaker can deploy models to various targets, including real-time endpoints, serverless inference, asynchronous inference, and batch transform, not only EC2 instances.

Answer

On-premises training only

Answer

Can only deploy models to EC2 instances

Question 7

Refer to the exhibit. A data scientist ran a training job on Amazon SageMaker and it failed. Which action should the data scientist take FIRST to resolve the issue?

Accepted Answer

Use a different instance type that is available in the region. The error indicates that the requested instance type is not available in the current region due to capacity constraints. The first step is to switch to a different instance type that is available in the same region, as this is the quickest and most direct way to resolve the provisioning failure without requiring service limit increases or changing regions.

Answer

Request a service limit increase for the instance type

Answer

Use a different AWS region

Answer

Enable spot training

Question 8

Refer to the exhibit. A developer wants to ensure the notebook instance can access the internet to download packages. Which property configuration ensures this?

Accepted Answer

DirectInternetAccess: Enabled. Setting `DirectInternetAccess: Enabled` on an AWS SageMaker notebook instance allows it to access the internet through a VPC with a Network Address Translation (NAT) gateway or via the public internet if the instance is not in a VPC. This configuration is required to download packages from external repositories like PyPI or conda.

Answer

VolumeSizeInGB: 5

Answer

InstanceType: ml.t2.medium

Answer

The resource type AWS::SageMaker::NotebookInstance

Question 9

Refer to the exhibit. A user tries to create a training job that reads data from a bucket named 'my-bucket'. The job fails with an access denied error. What is the most likely cause?

Accepted Answer

The s3:GetObject permission is missing the bucket ARN for listing objects. The error 'access denied' when reading data from S3 typically indicates that the IAM role used by SageMaker lacks the necessary permissions. Specifically, the s3:ListBucket permission is required to list objects in the bucket, and the s3:GetObject permission is required to retrieve them. Without the s3:ListBucket permission (or with an incomplete resource ARN), the job cannot enumerate objects, leading to the access denied error. Option C correctly identifies the missing bucket ARN for listing objects, which corresponds to the missing s3:ListBucket action.

Answer

The s3:GetObject action is restricted to a specific bucket but the bucket ARN is incorrect

Answer

The sagemaker:CreateTrainingJob action is not allowed for the specific instance type

Answer

The IAM role does not have permission to write to the bucket

Question 10

A developer needs to preprocess a dataset consisting of customer reviews for sentiment analysis. Which text preprocessing technique is most likely to improve model accuracy?

Accepted Answer

All of the above. All three listed techniques—stemming, removing stop words, and lowercasing—are standard text preprocessing steps that collectively improve model accuracy for sentiment analysis. Stemming reduces words to root forms to consolidate similar meanings, removing stop words eliminates noise from high-frequency but low-information tokens, and lowercasing normalizes case variations. Together, they reduce the feature space and help the model focus on sentiment-bearing terms, leading to better generalization and accuracy.

Answer

Stemming

Answer

Removing stop words

Answer

Lowercasing

Question 11

An organization wants to use Amazon Rekognition to analyze images of people for a security application. They must comply with GDPR. What is the best practice?

Accepted Answer

Ensure all images are anonymized before analysis. GDPR requires that personal data, including facial images, be processed lawfully and with appropriate safeguards. Anonymizing images before analysis with Amazon Rekognition ensures that the data cannot be linked back to an identifiable person, thereby reducing GDPR compliance risk. This aligns with the principle of data minimization and privacy by design.

Answer

Store images indefinitely for audit

Answer

Use celebrity recognition

Answer

Use face detection only

Question 12

In a binary classification problem, the model predicts majority class for all inputs. What is this issue called?

Accepted Answer

High bias. When a model predicts the majority class for all inputs, it indicates that the model is too simplistic and fails to capture the underlying patterns in the data. This is a classic symptom of high bias, where the model makes strong assumptions about the data distribution, leading to systematic underperformance on the minority class. In machine learning, high bias often results from an overly simple algorithm or insufficient model capacity, causing the model to underfit the training data.

Answer

Overfitting

Answer

High variance

Answer

Underfitting

Question 13

A SageMaker endpoint is configured with automatic scaling. The model's inference time is 50ms, and traffic increases gradually. What scaling metric should be used to add instances before latency increases?

Accepted Answer

Invocations per instance. D is correct because 'Invocations per instance' is a custom metric that directly measures the number of inference requests each instance is handling. By setting a target value for this metric, the scaling policy can proactively add instances when the per-instance request count approaches a threshold, preventing latency increases before they occur. This is the recommended approach for SageMaker endpoints with gradual traffic increases, as it anticipates demand rather than reacting to latency spikes.

Answer

Memory utilization

Answer

Concurrent requests

Answer

CPU utilization

Question 14

A company is training a large language model using Amazon SageMaker. The training job fails with the error 'OutOfMemory'. They are using a single ml.p3.2xlarge instance. The dataset is 50GB and the model is 2GB. The training script uses standard data loading. Which action should they take to resolve the issue?

Accepted Answer

Increase the instance type to ml.p3.16xlarge. The error 'OutOfMemory' indicates that the ml.p3.2xlarge instance (with 16 GB GPU memory) cannot hold both the 2 GB model and the 50 GB dataset during training. Increasing the instance type to ml.p3.16xlarge provides 64 GB GPU memory, which is sufficient to accommodate the model and dataset without memory pressure. This directly resolves the resource constraint.

Answer

Train the model using Spot instances

Answer

Reduce the batch size

Answer

Use SageMaker's Pipe mode for data loading

Question 15

A deployed model on an Amazon SageMaker endpoint is experiencing high inference latency (average 500ms) during peak hours. The model is a deep neural network with 10 million parameters. The endpoint uses a single ml.c5.xlarge instance. The company wants to reduce latency to under 200ms without retraining or changing the model architecture. Which action should they take?

Accepted Answer

Use SageMaker Neo to compile and optimize the model. SageMaker Neo compiles trained models into an optimized format for the target hardware, reducing inference latency without altering the model architecture. For a deep neural network with 10 million parameters on a CPU instance, Neo applies hardware-specific optimizations like operator fusion and memory layout tuning, which can significantly lower latency. This directly addresses the requirement to reduce latency from 500ms to under 200ms without retraining or changing the model.

Answer

Enable automatic scaling to add more instances

Answer

Switch to a GPU-based instance type like ml.p2.xlarge

Answer

Deploy the model on a multi-model endpoint