MLS-C01 Modeling Practice Test 42 — 15 Questions

Question 1

A company is using Amazon SageMaker to deploy a model for real-time inference. The model receives requests that are small but arrive in bursts. The data scientist wants to minimize latency and cost. Which deployment option is MOST appropriate?

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

Accepted Answer

Use Amazon SageMaker Serverless Inference. Amazon SageMaker Serverless Inference is the most appropriate option because it automatically scales compute resources based on request volume, charges only for the compute time used during inference (per-millisecond billing), and has no idle costs. This matches the bursty, small-request pattern perfectly, minimizing both latency and cost without requiring manual instance management.

Answer

Use a real-time endpoint with a single instance

Answer

Use a multi-model endpoint with auto-scaling

Answer

Use a batch transform job triggered by a schedule

Question 2

A data scientist is tuning a gradient boosting model using Amazon SageMaker Automatic Model Tuning. The objective metric is AUC. The training job converges quickly but the final model has low AUC on the validation set. Which hyperparameter should the data scientist adjust to improve validation AUC?

Accepted Answer

Decrease the learning rate and increase the number of rounds. Decreasing the learning rate and increasing the number of rounds is the correct approach because a low learning rate forces the model to take smaller steps toward the optimum, reducing overfitting and allowing more trees to contribute to the ensemble. This combination often improves generalization and validation AUC when the training job converges too quickly, indicating that the model is overfitting or underfitting due to aggressive learning.

Answer

Increase the subsample ratio of training data

Answer

Increase the learning rate

Answer

Increase the maximum depth of trees

Question 3

A data scientist is using Amazon SageMaker to train a deep learning model with a large dataset. The training job fails with a 'CUDA out of memory' error. What is the MOST efficient way to resolve this issue?

Accepted Answer

Reduce the batch size. The 'CUDA out of memory' error occurs when the GPU's memory is insufficient to hold the model parameters, gradients, optimizer states, and the current batch of data. Reducing the batch size decreases the memory footprint per training step, allowing the model to fit within the available GPU memory without requiring a more expensive instance or sacrificing GPU acceleration.

Answer

Switch to a CPU-only instance

Answer

Use a larger instance type with more GPUs

Answer

Increase the batch size

Question 4

A company is building a recommendation system using matrix factorization. The training data contains user-item interactions. The model performs well on the training set but poorly on the test set. Which regularization technique should be applied to improve generalization?

Accepted Answer

Add L2 regularization to the user and item latent factors. L2 regularization (weight decay) penalizes large values in the user and item latent factor matrices, which helps prevent overfitting by encouraging the model to learn smoother, more generalizable representations. This is the standard regularization technique used in matrix factorization for collaborative filtering, as it directly controls the magnitude of the latent vectors without inducing sparsity.

Answer

Add L1 regularization to the user and item latent factors

Answer

Apply dropout to the latent factors during training

Answer

Use batch normalization on the factors

Question 5

Which TWO actions can help reduce overfitting in a decision tree model? (Choose 2.)

Accepted Answer

Prune the tree after training. Option A, pruning the tree after training, is correct because post-pruning (e.g., cost-complexity pruning with ccp_alpha in scikit-learn) removes branches that add little predictive power, reducing variance and thus overfitting. Option C, setting a minimum number of samples per leaf (min_samples_leaf), is correct because it prevents the tree from creating leaves that fit noise from very few samples, forcing more generalizable splits. Option B is wrong because increasing max_depth lets the tree grow deeper and fit training noise more closely, worsening overfitting. Option D is wrong because increasing the number of features considered at each split (max_features) tends to make splits more specific to training data, not less. Option E is wrong because training on all data without a validation set removes the ability to detect or tune against overfitting.

Answer

Increase the maximum depth of the tree

Answer

Increase the number of features considered at each split

Answer

Use all training data without validation

Question 6

Which THREE factors should be considered when selecting the appropriate algorithm for a regression problem? (Choose 3.)

Accepted Answer

The number of features relative to the number of samples. Option A is correct because the ratio of features to samples directly affects algorithm choice: with many features relative to samples, high-variance models like unregularized linear regression or deep neural networks overfit, so regularized methods (Ridge/Lasso/Elastic Net) or dimensionality reduction become preferable, while with abundant samples flexible models are viable. Option B is correct because interpretability requirements constrain the model family: if stakeholders need explainable coefficients, linear regression or decision trees are appropriate, whereas black-box models like gradient boosting or neural networks may be unacceptable despite higher accuracy. Option C is correct because non-linear relationships determine whether a linear model suffices or whether polynomial features, kernel methods (e.g., SVR with RBF kernel), tree ensembles, or neural networks are needed to capture the underlying function. Option D is not a modeling factor—training time of day has no bearing on algorithm suitability, only on scheduling and resource availability. Option E is irrelevant; the color of a laptop has no technical relationship to regression algorithm selection.

Answer

The time of day the training will occur

Answer

The color of the data scientist's laptop

Question 7

Which TWO techniques are used for feature scaling? (Choose 2.)

Accepted Answer

Standardization (Z-score normalization). Standardization (Z-score normalization) (B) is a feature-scaling technique that transforms each feature by subtracting its mean and dividing by its standard deviation, producing values with mean 0 and standard deviation 1, which is essential for algorithms sensitive to feature magnitude such as SVM and k-means. Min-Max scaling (C) is also a feature-scaling technique that rescales each feature to a fixed range, typically [0, 1], using the formula (x - min) / (max - min), preserving the shape of the original distribution. One-hot encoding (A) is a categorical encoding method that creates binary columns, not a scaling technique, so it does not belong. Principal Component Analysis (D) is a dimensionality-reduction technique that projects data onto principal components, not a scaling method. Label encoding (E) assigns integer values to categorical labels and is an encoding technique, not feature scaling.

Answer

One-hot encoding

Answer

Principal Component Analysis (PCA)

Answer

Label encoding

Question 8

A data scientist is training a binary classification model on a highly imbalanced dataset (0.1% positive class). To improve recall, the team decides to use SageMaker's built-in XGBoost algorithm. Which parameter adjustment is most likely to increase recall without significantly sacrificing precision?

Accepted Answer

Set scale_pos_weight to the ratio of negative to positive samples. Setting scale_pos_weight to the ratio of negative to positive samples (approximately 999:1) tells XGBoost to assign a higher penalty to misclassifications of the minority positive class. This directly increases the gradient contribution from positive samples during training, which shifts the decision boundary to improve recall while maintaining a balance that avoids excessive false positives, thus preserving precision.

Answer

Increase max_depth from 5 to 10

Answer

Reduce num_round from 100 to 50

Answer

Increase subsample from 0.8 to 1.0

Question 9

A data scientist is using SageMaker to train a linear learner algorithm. After training, the evaluation shows that the model has high bias. Which action is most likely to reduce bias?

Accepted Answer

Add feature crosses for categorical variables. High bias indicates that the model is underfitting the data, meaning it is too simple to capture the underlying patterns. Adding feature crosses for categorical variables creates interaction features that allow the linear learner to model non-linear relationships, increasing model complexity and reducing bias. This is a standard technique in linear models to address underfitting without switching to a non-linear algorithm.

Answer

Increase the L2 regularization strength

Answer

Reduce the amount of training data

Answer

Remove some features that have low variance

Question 10

During training of a SageMaker built-in object detection algorithm, the loss is not decreasing after several epochs. Which troubleshooting step should be taken first?

Accepted Answer

Check whether the learning rate is appropriate. When the loss is not decreasing during training of a SageMaker built-in object detection algorithm, the most common cause is an inappropriate learning rate. A learning rate that is too high can cause the loss to oscillate or diverge, while one that is too low can cause the loss to plateau. Checking and adjusting the learning rate is the first troubleshooting step because it directly controls the step size of gradient updates and is a fundamental hyperparameter in optimization.

Answer

Increase the mini-batch size

Answer

Add more classes to the dataset

Answer

Increase the number of epochs

Question 11

A company is using SageMaker Autopilot to automatically build a binary classification model. After the AutoML job completes, the data scientist wants to understand which features are most important for the best candidate model. How can the scientist get feature importance?

Accepted Answer

Open the SageMaker Autopilot job details and view the 'Explainability' tab. SageMaker Autopilot automatically generates a 'Explainability' tab within the job details for the best candidate model. This tab uses SHAP (SHapley Additive exPlanations) values to provide feature importance, showing which features most influence the model's predictions. The data scientist can directly access this information without any additional configuration or re-running the model.

Answer

Re-run the best model using SageMaker built-in XGBoost with the 'feature_importance' hyperparameter

Answer

Check the CloudWatch Logs for the training job

Answer

Use SageMaker Ground Truth to label a new dataset

Question 12

A team is using SageMaker to train a custom PyTorch model on a large dataset (10 TB) stored in S3. The training job is repeatedly failing due to 'OutOfMemory' errors on the GPU. The team is using a single ml.p3.8xlarge instance. Which change is most likely to resolve the issue?

Accepted Answer

Reduce the batch size in the training script. The 'OutOfMemory' error on the GPU indicates that the model and its associated data exceed the available GPU memory. Reducing the batch size directly decreases the memory footprint per training step, allowing the model to fit within the GPU's memory limits. This is the most direct and effective fix for GPU OOM errors, as it reduces the amount of data processed simultaneously without changing the instance type or input mode.

Answer

Change the instance type to ml.p3.16xlarge (more GPUs)

Answer

Use managed spot training to reduce cost

Answer

Switch the input mode from Pipe to File

Question 13

Which TWO options are best practices for training machine learning models using SageMaker? (Choose TWO.)

Accepted Answer

Use incremental training when you have new data that is similar to the original training data. Option B is correct because SageMaker incremental training lets you reuse an existing trained model artifact and continue training it with new data that resembles the original dataset, avoiding a full retrain from scratch and saving time and compute. Option C is correct because SageMaker Managed Spot Training uses spare EC2 capacity at up to 90% lower cost, and SageMaker automatically manages checkpointing and job resumption when spot capacity is interrupted. Option A is not a best practice because the test set must remain held out to give an unbiased estimate of generalization; training on it causes data leakage and overly optimistic metrics. Option D is wrong because the largest instance type is not always optimal — it can waste money when the job is not compute-bound, and right-sizing via instance benchmarking is preferred. Option E is wrong because checkpointing after every epoch is not universally required; it adds S3 storage and I/O overhead and is mainly recommended for long-running or spot-training jobs.

Answer

Train the final model on the combined training and test sets to maximize data usage

Answer

Always use the largest possible instance type to minimize training time

Answer

Always enable checkpointing to save the model after every epoch

Question 14

Which THREE factors should be considered when choosing between SageMaker built-in algorithms and custom algorithms? (Choose THREE.)

Accepted Answer

Custom algorithms allow you to implement any architecture, including proprietary ones. Option A is correct because custom algorithms in SageMaker give you full flexibility to implement any model architecture, including proprietary or novel designs, by packaging your own training and inference code in a Docker container. Option B is correct because SageMaker built-in algorithms are engineered and tuned by AWS for scalable, distributed training on large datasets, often leveraging GPU instances and parallelization out of the box. Option E is correct because built-in algorithms expose a fixed set of predefined hyperparameters, which may not cover every tuning need or match a specialized use case, making custom algorithms preferable when finer control is required. Option C is incorrect because built-in algorithms support multiple input formats beyond CSV and JSON, such as RecordIO-Protobuf and Parquet, depending on the algorithm. Option D is incorrect because SageMaker built-in algorithms do support popular frameworks like PyTorch and TensorFlow through separate framework containers, and custom algorithms do require bringing your own container, but the claim about built-in algorithms not supporting PyTorch is false.

Answer

Built-in algorithms can only be used with CSV and JSON formats

Answer

Custom algorithms require you to bring your own Docker container, but SageMaker built-in algorithms do not support frameworks like PyTorch

Question 15

Which TWO actions can help reduce overfitting when training a model on SageMaker? (Choose TWO.)

Accepted Answer

Use early stopping based on validation error. Option B is correct because early stopping monitors a validation metric (e.g., validation error via the SageMaker EarlyStopping callback or the XGBoost early_stopping_rounds parameter) and halts training once the validation error stops improving, preventing the model from continuing to fit noise in the training set. Option E is correct because adding L1 regularization (in XGBoost, the alpha hyperparameter) penalizes large weights and drives some coefficients to zero, which constrains model complexity and reduces overfitting. Option A is wrong because reducing training data generally increases variance and makes overfitting worse, not better. Option C is wrong because increasing tree depth raises model capacity, allowing the model to memorize training data and overfit more. Option D is wrong because more training epochs let the model fit the training set even more closely, typically worsening overfitting.

Answer

Reduce the amount of training data

Answer

Increase the maximum depth of the trees in XGBoost

Answer

Increase the number of training epochs