MLS-C01 Modeling Practice Test 26 — 15 Questions

Question 1

A team is using Amazon SageMaker to train a deep learning model for image classification. The training job is taking too long, and they want to reduce training time without sacrificing model accuracy. Which approach is most effective?

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

Accepted Answer

Use transfer learning with a pre-trained model and fine-tune on the target dataset. Transfer learning with a pre-trained model leverages weights already learned on a large dataset (e.g., ImageNet) and fine-tunes them on the target dataset, dramatically reducing the number of epochs and compute needed to reach high accuracy. This is the most effective approach because the model starts from a strong feature representation rather than random initialization. It preserves or improves accuracy while cutting training time.

Answer

Reduce the batch size

Answer

Reduce the number of training epochs

Answer

Reduce the image resolution

Question 2

A machine learning engineer is tuning hyperparameters for a gradient boosting model using Amazon SageMaker Automatic Model Tuning. The objective metric is validation accuracy. After several tuning jobs, the best accuracy achieved is 0.85, but the engineer suspects the model is overfitting. Which hyperparameter adjustment is most likely to reduce overfitting?

Accepted Answer

Increase the regularization parameter (e.g., lambda or alpha). Increasing the regularization parameter (lambda for XGBoost, alpha for other implementations) penalizes model complexity, which reduces overfitting by discouraging the model from fitting noise in the training data. In SageMaker Automatic Model Tuning, this is a common adjustment when validation accuracy plateaus while training accuracy remains high.

Answer

Increase the maximum depth of trees

Answer

Increase the subsample ratio

Answer

Increase the learning rate

Question 3

A data scientist needs to choose an algorithm for a regression problem with 50 features and 1 million training examples. The model must be interpretable and the training data fits in memory. Which algorithm is most appropriate?

Accepted Answer

Linear regression. Linear regression is inherently interpretable — coefficients directly show the effect of each feature on the target — and it scales efficiently to 1 million rows with 50 features when the data fits in memory. It satisfies both the interpretability requirement and the computational constraints. More complex models like XGBoost may perform better but sacrifice the transparency the question demands.

Answer

Principal Component Analysis (PCA)

Answer

XGBoost

Answer

k-Nearest Neighbors

Question 4

A company is building a sentiment analysis model for customer reviews. The dataset includes 10,000 positive and 10,000 negative reviews. The data scientist splits the data into 70% training, 15% validation, and 15% test sets. After training, the model achieves 99% accuracy on training set but only 82% on validation set. What is the most likely issue?

Accepted Answer

The model is overfitting. A large gap between training accuracy (99%) and validation accuracy (82%) is the classic signature of overfitting — the model has memorized the training data, including noise, and fails to generalize to unseen data. With 20,000 balanced examples, the dataset is not imbalanced, and the model is clearly not underfitting since training performance is near-perfect. The remedy is regularization, dropout, early stopping, or more training data.

Answer

There is data leakage from validation to training

Answer

The dataset is imbalanced

Answer

The model is underfitting

Question 5

A company uses a linear regression model to predict house prices. The model's R-squared is 0.95 on the training set but 0.60 on the test set. Which of the following is the most likely cause?

Accepted Answer

Overfitting. A large gap between training R-squared (0.95) and test R-squared (0.60) is the classic signature of overfitting: the model has learned the training data too well, including noise, and fails to generalize to unseen data. Underfitting would show low performance on both sets, not a large disparity.

Answer

Underfitting

Answer

Multicollinearity among features

Answer

Data leakage

Question 6

A data scientist needs to perform feature scaling for a dataset containing numerical features with different units (e.g., age in years and income in dollars). Which scaling method is most appropriate when the algorithm assumes data is normally distributed?

Accepted Answer

Standardization (Z-score normalization). Standardization (Z-score normalization) transforms features to have mean 0 and standard deviation 1, which is suitable for algorithms that assume normally distributed data. Option B (Log transformation) is used to reduce skewness, not for scaling when normality is already assumed. Option C (Min-Max scaling) scales to a fixed range (e.g., 0-1) but does not preserve normal distribution properties. Option D (Robust scaling) uses median and IQR, which is robust to outliers but does not assume normality.

Answer

Log transformation

Answer

Min-Max scaling

Answer

Robust scaling (using median and IQR)

Question 7

Which THREE of the following are appropriate methods to reduce overfitting in a decision tree model?

Accepted Answer

Prune the tree after training. Option C (Prune the tree after training) is correct because post-pruning removes branches that add little predictive power, reducing variance and thus overfitting. Option D (Set a minimum number of samples required to split an internal node) is correct because requiring more samples per split prevents the tree from creating splits based on noise in small sample subsets. Option E (Limit the maximum depth of the tree) is correct because capping depth restricts the tree's ability to memorize training data through overly deep, highly specific splits. Option A is incorrect because increasing the number of features considered per split tends to make the tree more expressive and can worsen overfitting. Option B is incorrect because increasing maximum depth allows the tree to grow deeper and fit noise, which increases overfitting.

Answer

Increase the number of features considered for each split

Answer

Increase the maximum depth of the tree

Question 8

Which TWO of the following are true about the bias-variance tradeoff?

Accepted Answer

High variance can cause overfitting. Option C is correct because high variance means the model's predictions change substantially with different training samples, so it fits noise in the training data and generalizes poorly, which is the definition of overfitting. Option D is correct because high bias means the model makes strong, oversimplified assumptions that it cannot capture the underlying pattern, producing systematic error on both training and test data, which manifests as underfitting. Option A is wrong because bagging (bootstrap aggregating) averages multiple models to reduce variance, not increase it. Option B is wrong because simple models typically have high bias and low variance, not high variance. Option E is wrong because high variance models are typically complex and flexible, not too simple.

Answer

Ensemble methods like bagging increase variance

Answer

Simple models tend to have high variance

Answer

High variance models are typically too simple

Question 9

A team deployed a SageMaker endpoint with the configuration shown in the exhibit. During a traffic spike, the endpoint becomes unresponsive. Which change to the endpoint configuration would best improve availability?

Accepted Answer

Configure auto-scaling for the endpoint. The endpoint is unresponsive during a traffic spike because the fixed instance count cannot handle the increased load. SageMaker endpoints support automatic scaling, which dynamically adjusts the number of instances based on metrics like InvocationsPerInstance or CPUUtilization. Configuring auto-scaling allows the endpoint to add instances during spikes, thereby maintaining availability and performance. This is the best change because it directly addresses the root cause—insufficient capacity under variable load—without over-provisioning.

Answer

Reduce the initial instance count to 0 and use on-demand invocation

Answer

Add a second production variant with the same model

Answer

Change the instance type to ml.m5.xlarge

Question 10

A machine learning engineer is using Amazon SageMaker to train a deep learning model. The training job is taking longer than expected. The engineer notices that the GPU utilization is low (around 30%) while CPU utilization is high. Which action is most likely to improve training speed?

Accepted Answer

Increase the number of data loading workers. Low GPU utilization with high CPU utilization indicates the data pipeline (host-side preprocessing, augmentation, disk I/O) cannot feed the GPU fast enough. Increasing the number of data loading workers parallelizes data fetching and preprocessing, raising GPU utilization and reducing training time.

Answer

Use a smaller instance type with fewer GPUs

Answer

Decrease the number of data loading workers

Answer

Increase the batch size

Question 11

A data scientist is building a regression model to predict house prices. The dataset includes features such as square footage, number of bedrooms, and location. After training a linear regression model, the scientist notices that the residuals have a pattern: they increase as the predicted value increases. Which action is most appropriate?

Accepted Answer

Apply a log transformation to the target variable. A residual pattern that fans out (increases) as predicted values increase indicates heteroscedasticity and a right-skewed target, which violates linear regression's constant-variance and normality assumptions. Applying a log transformation to the target compresses large values, stabilizes variance, and often linearizes the relationship, directly addressing the pattern. This is the standard remedy for multiplicative or skewed regression targets like house prices.

Answer

Remove outliers from the dataset

Answer

Use Ridge regression instead of linear regression

Answer

Add polynomial features to the model

Question 12

A machine learning team is using Amazon SageMaker to train a large language model. The training script uses PyTorch and the model requires significant memory. The team wants to use model parallelism across multiple GPUs. Which SageMaker feature should they use?

Accepted Answer

SageMaker model parallelism library. The SageMaker model parallelism library is specifically designed to split a single large model across multiple GPUs, enabling training of models that don't fit in a single GPU's memory. It provides automatic and manual partitioning strategies for PyTorch and TensorFlow, making it the correct choice for model parallelism across multiple GPUs. This library handles the communication and synchronization needed when model layers are distributed across devices.

Answer

SageMaker Distributed Training

Answer

SageMaker Horovod

Answer

SageMaker Debugger

Question 13

A machine learning team is using Amazon SageMaker to train a model using a custom Docker container. The training job fails with an error: 'Unable to write to /opt/ml/model'. The container does not have root access. What is the most likely cause?

Accepted Answer

The container process does not have write permission to /opt/ml/model. Amazon SageMaker mounts a writable volume at /opt/ml/model inside the training container, but the directory is owned by root with permissions that may not allow non-root users to write. Since the container runs as a non-root user, it lacks the necessary write permission, causing the 'Unable to write to /opt/ml/model' error. The fix is to ensure the Dockerfile creates the directory with appropriate ownership or the training script writes to a different location that is writable, such as /opt/ml/output or /tmp, but the model artifacts must ultimately be saved to /opt/ml/model for SageMaker to capture them.

Answer

The /opt/ml/model directory does not exist in the container

Answer

The container is using an unsupported operating system

Answer

The container does not have internet access

Question 14

A data scientist is building a classification model to predict customer churn. The dataset has 10,000 samples with 100 features. After training a logistic regression model, the scientist observes that the model has high variance (overfitting). Which technique can reduce overfitting?

Accepted Answer

Use L2 regularization (Ridge). L2 regularization (Ridge) adds a penalty proportional to the square of the coefficients to the loss function, which shrinks coefficients and reduces model variance, thereby mitigating overfitting. It is a standard technique for high-variance models like logistic regression with many features. By penalizing large weights, it improves generalization on unseen data.

Answer

Remove the regularization term

Answer

Add polynomial features

Answer

Use a smaller learning rate

Question 15

Which TWO of the following are valid approaches to handle missing values in a dataset for a machine learning model?

Accepted Answer

Impute missing values with the mean of the column. Option B is correct because mean imputation replaces missing numeric entries with the column's average, a standard and simple technique that preserves the row count and avoids discarding data. Option C is correct because listwise deletion (dropping rows containing missing values) is a valid, widely used approach when the missingness is random and the remaining data is sufficient. Option A is not a standard or reliable missing-value handling method; neural networks can model data but are not a defined imputation technique and would themselves require complete training data. Option D is incorrect because standardization rescales feature values and does nothing to fill or remove missing entries. Option E is incorrect because one-hot encoding is for converting categorical variables into binary vectors, not for handling missing values.

Answer

Use a neural network to predict missing values

Answer

Standardize the features to handle missing values

Answer

Apply one-hot encoding to convert missing values