AIF-C01 Fundamentals of AI and ML — 10 Questions

Question 1

A company is training a deep learning model on Amazon SageMaker using a custom Docker container. The training job fails with the error 'CannotStartContainerError: API error (500): failed to create shim task'. The team verifies that the container image is compatible with the selected instance type. What is the most likely cause of this error?

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

Accepted Answer

The instance type does not have enough memory for the container. The error 'CannotStartContainerError: API error (500): failed to create shim task' typically occurs when the Docker container cannot be initialized due to resource constraints, most commonly insufficient memory on the selected instance type. Even if the container image is compatible with the instance, the container's memory request may exceed the available memory, causing the container runtime (containerd) to fail when creating the shim task. This is a known issue in SageMaker when the training job's resource requirements are not aligned with the instance's capacity.

Answer

The training data is stored in the wrong S3 bucket

Answer

The container image does not have the correct entry point

Answer

The GPU drivers are outdated

Question 2

A machine learning engineer is using Amazon SageMaker to train a model and wants to automatically stop the training job if the loss does not improve for 10 consecutive epochs. Which SageMaker feature should be used?

Accepted Answer

SageMaker built-in algorithms with early stopping. Amazon SageMaker built-in algorithms support early stopping, which allows you to automatically terminate a training job when a specified metric, such as loss, stops improving for a defined number of consecutive epochs. This feature is configured directly in the algorithm's hyperparameters (e.g., `early_stopping_patience` for the XGBoost algorithm) and helps save compute time and cost by preventing overfitting.

Answer

SageMaker Training Compiler

Answer

SageMaker Debugger

Answer

SageMaker Experiments

Question 3

During a SageMaker training job, the data scientist observes that the loss is not decreasing after the initial few epochs. The model is a deep neural network with ReLU activations. Which hyperparameter adjustment is most likely to help?

Accepted Answer

Reduce the learning rate. When loss plateaus after a few epochs with ReLU activations, the model is likely stuck in a region where gradients are small (e.g., near a local minimum or plateau). Reducing the learning rate allows the optimizer to take smaller steps, which can help it navigate out of flat regions and continue decreasing the loss. This is a standard technique to improve convergence when training stalls.

Answer

Increase the number of epochs

Answer

Increase the learning rate

Answer

Decrease the batch size

Question 4

Which TWO factors should be considered when choosing between a CPU-based instance and a GPU-based instance for training a machine learning model on Amazon SageMaker? (Choose two.)

Accepted Answer

The size of the dataset. Option C (the size of the dataset) is correct because dataset scale directly drives the compute and memory demands of training: very large datasets with heavy matrix operations benefit from GPU parallelism, while small datasets often train efficiently on CPU instances, making dataset size a key cost/performance factor in choosing between CPU and GPU. Option E (the type of model architecture, e.g., CNN vs. linear regression) is correct because architecture determines whether the workload is dominated by massively parallel tensor math (CNNs, transformers, deep networks) that GPUs accelerate, or by simpler, largely sequential operations (linear regression, small tree-based models) where CPU instances are sufficient and more cost-effective. Option A (number of layers) is not a primary selection factor on its own, since depth matters only insofar as it reflects the architecture and compute pattern already captured by option E. Option B (AWS Region) affects availability, pricing, and latency but not the fundamental CPU-versus-GPU compute decision. Option D (choice of hyperparameter optimizer) influences convergence behavior and tuning, not the underlying hardware class needed for training.

Answer

The number of layers in the model

Answer

The AWS Region

Answer

The choice of hyperparameter optimizer

Question 5

Refer to the exhibit. A data scientist attaches the above IAM policy to a SageMaker notebook instance role. The notebook is in the same AWS account as the S3 bucket. When trying to read a file from 's3://my-bucket/training/data.csv', the data scientist gets an Access Denied error. What is the most likely cause?

Accepted Answer

The S3 bucket is in a different Region. The Access Denied error is most likely due to the S3 bucket being in a different Region. While IAM policies are global, bucket policies can include conditions that restrict access based on the request's Region. If the bucket policy denies requests from the Region where the SageMaker notebook is running, the read operation fails. The other options are incorrect: file name spaces cause a different error (NoSuchKey), s3:ListBucket is not required for direct object reads with GetObject, and allowing s3:PutObject does not cause Access Denied for reads.

Answer

The file name contains spaces

Answer

The policy does not grant 's3:ListBucket' permission

Answer

The policy allows 's3:PutObject' which is not needed

Question 6

Refer to the exhibit. A data scientist is training a neural network model on SageMaker. The training log shows the loss values per epoch. Which issue is most likely occurring?

Accepted Answer

The model is overfitting. The training log shows loss values decreasing on the training set but increasing or plateauing on the validation set, which is a classic sign of overfitting. Overfitting occurs when the model learns noise and specific patterns in the training data too well, failing to generalize to unseen data. In SageMaker, monitoring both training and validation loss curves is critical to detect this issue early.

Answer

The number of epochs is insufficient

Answer

The dataset is too small

Answer

The learning rate is too low

Question 7

A company is deploying a machine learning model for real-time fraud detection. The model must make predictions with latency under 10 milliseconds. The data scientist trained a gradient boosting model that achieves high accuracy but has inference latency of 50 milliseconds. The team has access to a larger instance type with more CPU cores. Which approach should the data scientist take to reduce inference latency while maintaining accuracy?

Accepted Answer

Deploy the model on a larger instance with more CPU cores.. Increasing the number of CPU cores allows the gradient boosting model to parallelize tree evaluation across multiple cores, reducing inference latency. Since the model is already trained and accurate, this hardware scaling directly addresses the 50 ms bottleneck without altering the model's structure or accuracy.

Answer

Switch to batch inference and run predictions every 100 milliseconds.

Answer

Reduce the maximum tree depth and retrain the model.

Answer

Apply post-training pruning to remove redundant trees.

Question 8

A data scientist is preparing data for a machine learning model. What is the purpose of splitting the data into training, validation, and test sets?

Accepted Answer

To evaluate model generalization. The test set is held out entirely from the training process and is used only once at the end to evaluate the model's performance on unseen data. This provides an unbiased estimate of how well the model generalizes to new, real-world data, which is the ultimate goal of supervised learning.

Answer

To tune hyperparameters

Answer

To balance class distributions

Answer

To prevent overfitting during training

Question 9

A team trains a model using Amazon SageMaker built-in XGBoost. After training, they want to evaluate feature importance. Which SageMaker feature allows them to view this?

Accepted Answer

SageMaker Debugger. SageMaker Debugger provides built-in monitoring and visualization capabilities, including the ability to capture feature importance metrics (e.g., gain, cover, weight) from XGBoost training jobs. It automatically saves these metrics to Amazon S3 and allows you to view them through the SageMaker Studio Debugger dashboard or by querying the saved tensors, enabling direct evaluation of feature importance without additional custom code.

Answer

SageMaker Experiments

Answer

SageMaker Autopilot

Answer

SageMaker Model Monitor

Question 10

A data scientist is working with a dataset that contains both numerical and categorical features. Which algorithm is commonly used for regression tasks in AWS SageMaker?

Accepted Answer

Linear Learner. Linear Learner (option B) is the correct choice because it is a supervised learning algorithm in AWS SageMaker specifically designed for both regression and classification tasks. It can handle datasets with mixed numerical and categorical features (after appropriate encoding) and provides built-in mechanisms for training linear models, including automatic model tuning and distributed training. Note that option D is a duplicate of option B and is not considered the correct answer in this context.

Answer

K-Means

Answer

BlazingText

Answer

Linear Learner