MLS-C01 Modeling Practice Test 12 — 15 Questions

Question 1

A company is using SageMaker to deploy a real-time inference endpoint for a natural language processing model. The model receives input text and returns predictions. The data scientist notices that the endpoint latency increases significantly under load. Which design change would MOST effectively reduce latency?

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

Accepted Answer

Use an inference pipeline to combine preprocessing and model inference. An inference pipeline in SageMaker allows you to chain preprocessing logic directly with the model inference within the same endpoint container. This eliminates the need for separate Lambda functions or client-side preprocessing, which reduces network round-trips and serialization overhead, thereby lowering latency under load.

Answer

Enable data capture for monitoring

Answer

Switch to batch transform for real-time predictions

Answer

Increase the number of instances behind the endpoint

Question 2

A data scientist is training a deep learning model for object detection. The training loss decreases rapidly in the first few epochs but then plateaus at a high value. The validation loss starts increasing after a few epochs. Which adjustment is MOST likely to improve generalization?

Accepted Answer

Implement early stopping with a patience parameter. The described behavior—training loss plateauing at a high value while validation loss increases—is a classic sign of overfitting. Early stopping with a patience parameter halts training when validation performance stops improving, preventing the model from memorizing noise and thus improving generalization. This directly addresses the overfitting without altering the model architecture or data distribution.

Answer

Add more convolutional layers

Answer

Use more aggressive data augmentation

Answer

Increase the learning rate

Question 3

A data scientist is building a model to predict insurance claim amounts. The target variable is right-skewed with many small claims and a few very large claims. The scientist wants to minimize the impact of outliers. Which loss function or transformation is MOST appropriate?

Accepted Answer

Apply a log transformation to the target variable. Applying a log transformation to the target variable reduces skewness and mitigates the impact of outliers by compressing the scale of large values. This makes the distribution more symmetric and suitable for models like linear regression. Option A (mean squared error) is sensitive to outliers. Option B (quantile loss) predicts the median, which is robust but not typical for mean prediction. Option C (Poisson loss) is designed for count data, not continuous skewed targets. Option D (log transformation) is the standard approach for right-skewed continuous targets.

Answer

Use mean squared error loss without any transformation

Answer

Use quantile loss to predict the median

Answer

Use Poisson loss assuming the target follows a Poisson distribution

Question 4

A data scientist is using Amazon SageMaker to train a deep learning model. The training job is taking too long. Which THREE actions can reduce training time?

Accepted Answer

Use incremental training to continue from a previous model. Option A is correct because incremental training lets the job resume from an existing model's artifacts instead of retraining from scratch, so the total training time is reduced. Option C is correct because Pipe input mode streams data directly from Amazon S3 into the training container, avoiding the latency of downloading the full dataset to local storage before training begins. Option E is correct because GPU instance types (e.g., ml.p3 or ml.p4 families) provide hardware acceleration for deep learning matrix operations, which speeds up training compared to CPU-only instances. Option B is not correct because Spot Instances reduce cost, not training time, and can even be interrupted and restarted. Option D is not correct because decreasing batch size generally increases the number of iterations needed per epoch and does not reliably reduce overall training time.

Answer

Use Spot Instances to reduce cost

Answer

Decrease the batch size to reduce memory usage

Question 5

A company is building a sentiment analysis model for customer reviews. The dataset is balanced with 10,000 positive and 10,000 negative reviews. The model achieves 95% accuracy on the test set but fails to generalize to new reviews from a different product category. Which TWO techniques can improve generalization?

Accepted Answer

Use stratified k-fold cross-validation during training. Option B is correct because stratified k-fold cross-validation preserves the class ratio in every fold and evaluates the model across multiple train/validation splits, exposing variance and reducing the chance that performance is tied to one particular split, which directly improves generalization to unseen data. Option C is correct because L2 regularization adds a penalty proportional to the squared magnitude of the weights, shrinking them and limiting overfitting so the model does not memorize training-specific patterns that fail on a different product category. Option A is not among the marked answers; while more data can help, the scenario already has a balanced 20,000-review set and the failure is domain shift, so simply adding same-domain reviews does not guarantee better cross-category generalization. Option D is not marked because adding features such as review length and word count can introduce noise or spurious correlations rather than improve generalization. Option E is not marked because a more complex model with more layers increases capacity and typically worsens overfitting, the opposite of what is needed here.

Answer

Increase the training dataset size by collecting more reviews

Answer

Add more features like review length and word count

Answer

Use a more complex model with more layers

Question 6

Refer to the exhibit. A data scientist wants to use SageMaker to train a model using data stored in 'my-bucket'. The training job fails with an access denied error. What is the MOST likely cause?

Accepted Answer

The training job is not requesting server-side encryption with AES256. SageMaker training jobs require server-side encryption with AES256 when the S3 bucket uses default encryption with AES256. If the training job does not explicitly request SSE-S3 (AES256) in its S3 data source configuration, the S3 service denies access, resulting in an 'access denied' error even if the IAM role has full S3 permissions.

Answer

The bucket uses AWS KMS key encryption instead of AES256

Answer

The bucket name in the policy does not match the actual bucket name

Answer

The bucket is publicly accessible but the IAM role lacks permissions

Question 7

A machine learning team is deploying a real-time inference endpoint for a fraud detection model using Amazon SageMaker. The model requires low latency (<100 ms) and the team expects a steady stream of requests with occasional spikes. Which instance type and deployment strategy should they use to minimize cost while meeting latency requirements?

Accepted Answer

Use ml.c5 instances with a single endpoint and auto-scaling with a minimum instance count to handle baseline traffic.. Ml.c5 instances are compute-optimized for low-latency inference, and auto-scaling with a minimum instance count ensures pre-warmed capacity to handle steady traffic with spikes without cold starts, meeting the <100 ms requirement cost-effectively. This combination avoids over-provisioning while ensuring consistent performance.

Answer

Use ml.p3 instances with a multi-model endpoint.

Answer

Use AWS Lambda with a container image for serverless inference.

Answer

Use ml.m5 instances with a production variant and auto-scaling.

Question 8

A data scientist is training a deep learning model on Amazon SageMaker using a PyTorch estimator. The training job runs on a single ml.p3.2xlarge instance but is taking too long. The scientist wants to reduce training time by using distributed data parallelism across multiple GPUs. Which change to the training script and SageMaker estimator is required?

Accepted Answer

Modify the script to use torch.nn.parallel.DistributedDataParallel and set instance_count to 2 in the estimator.. To achieve distributed data parallelism across multiple GPUs on multiple instances with PyTorch, you must modify the training script to use `torch.nn.parallel.DistributedDataParallel` (DDP), which handles gradient synchronization across nodes. Additionally, you must set `instance_count` to 2 (or more) in the SageMaker PyTorch estimator to launch multiple instances, each with its own GPU, enabling true multi-node distributed training.

Answer

Add the SageMaker distributed data parallelism configuration in the estimator and modify the script to use the SageMaker distributed library.

Answer

Change the framework to TensorFlow and use tf.distribute.MirroredStrategy with instance_count=2.

Answer

Modify the script to use torch.nn.DataParallel and keep instance_count as 1.

Question 9

A machine learning engineer is deploying a model to an Amazon SageMaker endpoint for real-time inference. The model is a large ensemble that requires 4 GB of memory. The engineer wants to minimize cost while ensuring the endpoint can handle up to 100 concurrent requests with a latency under 200 ms. Which instance configuration is most appropriate?

Accepted Answer

One ml.c5.xlarge instance with auto-scaling up to 2 instances.. The ml.c5.xlarge instance provides sufficient compute (4 vCPUs, 8 GB memory) for the 4 GB model, and auto-scaling up to 2 instances allows handling 100 concurrent requests with low latency while minimizing cost during low traffic. The ml.c5 family is optimized for compute-intensive inference, and auto-scaling ensures the endpoint scales out only when needed, avoiding over-provisioning.

Answer

Two ml.t3.medium instances behind a load balancer.

Answer

One ml.m5.2xlarge instance.

Answer

One ml.p3.2xlarge instance.

Question 10

A data scientist is tuning a gradient boosting model using Amazon SageMaker Automatic Model Tuning (AMT). Which THREE hyperparameters should the scientist consider tuning to reduce overfitting? (Select THREE.)

Accepted Answer

Learning rate (eta). Learning rate (eta) (B) is correct because a smaller eta shrinks each boosting step, forcing the model to learn more gradually and reducing the risk of fitting noise in the training data. Minimum child weight (min_child_weight) (C) is correct because raising it requires a larger sum of instance weights before a leaf can be created, which regularizes the tree and prevents overly specific splits. Gamma (minimum loss reduction) (D) is correct because it sets the minimum loss reduction needed to make a further split, so higher values prune splits that yield little gain and curb overfitting. Subsample ratio (A) and maximum depth (E) are not among the marked correct answers here, even though they can affect model complexity, so they should not be selected for this specific set.

Answer

Subsample ratio

Answer

Maximum depth (max_depth)

Question 11

Refer to the exhibit. The log shows the end of a successful SageMaker training job. However, the ML engineer cannot find the model artifacts in the specified S3 bucket. What is the most likely cause?

Accepted Answer

The IAM role used by the training job does not have permission to write to the S3 bucket.. The training job completed successfully, meaning the SageMaker training container executed without errors. However, if the model artifacts are not found in the specified S3 bucket, the most likely cause is that the IAM role associated with the training job lacks the necessary s3:PutObject permission for that bucket. SageMaker uses the role's credentials to write the output; without write access, the artifacts are silently dropped or fail to upload, even though the training code itself may have run to completion.

Answer

The S3 bucket does not exist.

Answer

The model artifacts were uploaded to a different S3 path.

Answer

The training job did not have network access to S3.

Question 12

Refer to the exhibit. An IAM policy is attached to a SageMaker notebook instance. A data scientist is trying to invoke the endpoint 'my-endpoint' from the notebook but receives an AccessDenied error. What is the likely cause?

Accepted Answer

The policy is not attached to the IAM role used by the notebook instance.. The error 'AccessDenied' when invoking a SageMaker endpoint from a notebook instance typically indicates that the IAM role attached to the notebook does not have the required permissions. The policy shown in the exhibit grants sagemaker:InvokeEndpoint for the specific endpoint ARN, but if the policy is not attached to the IAM role that the notebook instance is using, the role lacks the permission, resulting in the AccessDenied error. Attaching the policy to the correct IAM role resolves the issue.

Answer

The policy allows InvokeEndpoint only for endpoints with the exact ARN, but the endpoint ARN is different.

Answer

The policy uses a wildcard for CreateEndpoint, which is too permissive.

Answer

The policy does not allow sagemaker:CreateEndpoint for the specific endpoint.

Question 13

A company is deploying a real-time fraud detection model using Amazon SageMaker. The model must make predictions in under 100 milliseconds. The data scientist uses a pre-trained XGBoost model and deploys it to a SageMaker endpoint with an ml.c5.xlarge instance. After load testing, the average latency is 150 ms. Which action should the data scientist take to reduce latency?

Accepted Answer

Enable SageMaker Neo to compile the model for the target instance. SageMaker Neo optimizes trained models for the target hardware platform by compiling them into an efficient runtime. This reduces inference latency without changing the model architecture, making it ideal for meeting the sub-100ms requirement when the current latency is 150ms on an ml.c5.xlarge instance.

Answer

Reduce the number of trees in the XGBoost model

Answer

Deploy multiple instances behind a load balancer

Answer

Use a larger instance type to increase compute capacity

Question 14

A company wants to build a machine learning model to predict customer churn. The dataset includes customer demographics, usage patterns, and support interactions. The data is stored in Amazon S3. The data scientist needs to perform feature engineering, including creating aggregate features from support interactions and encoding categorical variables. Which AWS service is most suitable for building the feature engineering pipeline?

Accepted Answer

Amazon SageMaker Processing. Amazon SageMaker Processing is the most suitable service because it is purpose-built for data preprocessing and feature engineering within the SageMaker ecosystem. It allows you to run custom Python scripts (e.g., using pandas or PySpark) on managed infrastructure to create aggregate features from support interactions and encode categorical variables, and it integrates seamlessly with SageMaker for model training and deployment.

Answer

AWS Glue

Answer

Amazon EMR

Answer

AWS Batch

Question 15

A data scientist is assigned an IAM policy as shown. The data scientist attempts to create a SageMaker endpoint to deploy a model, but the request fails. What is the most likely reason?

Accepted Answer

The data scientist does not have permission to create an endpoint.. The IAM policy shown does not include the `sagemaker:CreateEndpoint` action, which is required to create a SageMaker endpoint. Even if the data scientist has permissions for other SageMaker actions like `CreateModel` or `CreateEndpointConfig`, the explicit absence of `CreateEndpoint` in the policy will cause the request to fail with an access denied error. AWS IAM policies must explicitly grant each action needed for the operation.

Answer

The data scientist does not have permission to upload the model to S3.

Answer

The data scientist does not have permission to create a training job.

Answer

The data scientist does not have permission to pass roles.