AI0-001 Machine Learning and Deep Learning Practice Test 3 — 15 Questions

Question 1

An e-commerce company deploys a deep learning model for product recommendation. After a new data pipeline is implemented, the model's online performance degrades: recall drops by 20% and the click-through rate decreases. The data scientists suspect data drift. They compare the distribution of the input features between the training data and recent production data. The Kolmogorov-Smirnov test shows significant differences for two numerical features (price and rating). The team also notices that the frequency of categorical feature 'category' has changed. Which of the following is the MOST appropriate first step? A. Immediately retrain the model on all available data including new production data. B. Roll back to the previous data pipeline and investigate the root cause of drift. C. Use feature selection to remove the drifting features and retrain. D. Implement a monitoring dashboard to track drift over time and set up alerts.

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

Accepted Answer

Roll back to the previous data pipeline and investigate the root cause of drift.. The most appropriate first step is to roll back to the previous data pipeline and investigate the root cause of drift. Rolling back restores the model's performance temporarily while the team diagnoses why the new pipeline introduced drift, such as changes in data collection or preprocessing. This prevents further degradation and allows for a controlled fix.

Answer

Implement a monitoring dashboard to track drift over time and set up alerts.

Answer

Use feature selection to remove the drifting features and retrain.

Answer

Immediately retrain the model on all available data including new production data.

Question 2

A data scientist is preparing a dataset for a binary classification neural network. The dataset contains both numerical and categorical features, and some rows have identical entries. Which TWO preprocessing steps are most essential to improve model performance and avoid overfitting?

Accepted Answer

Removing duplicate records. Removing duplicate records (A) is essential because identical rows can artificially inflate the importance of certain patterns, leading the model to memorize noise rather than generalize. This directly reduces overfitting by ensuring the training set reflects true data distribution. Scaling numerical features (B) to zero mean and unit variance (standardization) is critical for neural networks as it prevents features with larger magnitudes from dominating gradient updates, enabling faster convergence and stable training.

Answer

Increasing the batch size

Answer

Applying PCA for dimensionality reduction

Answer

Using dropout regularization in the model

Question 3

A junior data scientist is evaluating a binary classification model. The model's confusion matrix shows 90 true positives, 10 false negatives, 20 false positives, and 80 true negatives. What is the precision of the model?

Accepted Answer

0.818. Precision is the ratio of correctly predicted positive observations to the total predicted positives. Here, true positives are 90 and false positives are 20, so precision = 90/(90+20) = 0.818. This measures how accurate the model is when it predicts the positive class.

Answer

0.82

Answer

0.85

Answer

0.90

Question 4

A data scientist is training a logistic regression model to predict loan defaults. The dataset contains 10,000 records with 45 features, many of which are highly correlated. The model's test accuracy is 92%, but the training accuracy is 94%. The data scientist suspects that multicollinearity is affecting the model's interpretability. Which technique should be used to address multicollinearity while preserving the model's predictive power?

Accepted Answer

Ridge Regression (L2 regularization). Ridge regression is specifically designed to handle multicollinearity by adding an L2 penalty that constrains coefficient magnitudes. This reduces variance and stabilizes estimates when features are correlated, without removing features, so the model remains interpretable. The other methods either transform features (PCA) or perform feature selection (Lasso), which can harm interpretability or arbitrarily drop features.

Answer

Increasing the number of training epochs

Answer

Lasso Regression (L1 regularization)

Answer

Principal Component Analysis (PCA)

Question 5

A junior data analyst is evaluating a binary classification model that predicts whether a customer will churn. The model's confusion matrix shows 80 true positives, 10 false positives, 5 false negatives, and 100 true negatives. The analyst wants to calculate the precision of the model. Which value should the analyst compute?

Accepted Answer

0.89. Precision measures the proportion of positive identifications that were actually correct. It is calculated as true positives divided by the sum of true positives and false positives. With 80 true positives and 10 false positives, precision is 80/90 ≈ 0.89. This metric is crucial when the cost of false positives is high, such as in spam detection or churn prediction where false alarms are costly.

Answer

0.80

Answer

0.91

Answer

0.94

Question 6

A machine learning team is preparing a dataset of 50,000 labeled images for training a convolutional neural network. They want to apply data augmentation techniques that are label-preserving for this image classification task. Which TWO techniques should they use? (Choose two.)

Accepted Answer

Random horizontal flips of the images.. Label-preserving augmentation must transform inputs without changing the correct class. Horizontal flips and small translations both satisfy this for general image classification and are standard practice in convolutional pipelines. Rotating digits by 180 degrees swaps 6 and 9, occluding regions with a global mean can erase class-defining evidence, and randomizing labels injects noise rather than useful variation.

Answer

Random rotation of digits by 180 degrees in a handwritten digit dataset.

Answer

Assigning random class labels to a small percentage of the training images.

Answer

Replacing random image regions with the mean pixel value of the entire dataset.

Question 7

A machine learning team is preparing a dataset for training a model to predict housing prices. The dataset contains numerical features such as square footage, number of bedrooms, and age of the house, as well as categorical features such as neighborhood and style. The team wants to apply feature engineering techniques to improve model performance. Which TWO techniques are appropriate for handling the categorical features? (Choose two.)

Accepted Answer

Target encoding. One-hot encoding and target encoding are both appropriate for categorical features. One-hot encoding is standard for nominal data, creating binary indicators. Target encoding captures the relationship with the target variable and is effective for high-cardinality features. Both are widely used in practice, though target encoding requires careful implementation to prevent data leakage.

Answer

Principal Component Analysis (PCA)

Answer

Normalization (Min-Max scaling)

Answer

Label encoding

Question 8

A data scientist is building a classification model to detect fraudulent transactions. The dataset is highly imbalanced with only 1% fraudulent cases. Which approach should the scientist use to evaluate model performance most effectively?

Accepted Answer

F1 score. In highly imbalanced datasets like fraud detection (1% positive class), accuracy is misleading because a model that predicts all transactions as legitimate would achieve 99% accuracy yet fail to detect any fraud. The F1 score (harmonic mean of precision and recall) is the most effective metric because it balances both false positives and false negatives, providing a single score that reflects the model's ability to correctly identify the minority class without being skewed by class imbalance.

Answer

Accuracy

Answer

Recall

Answer

Precision

Question 9

A machine learning team is deploying a model that predicts customer churn. They notice that the model's predictions are highly sensitive to small changes in input features, leading to inconsistent outputs. Which technique should the team apply to improve model stability?

Accepted Answer

Regularization. Regularization (Option C) is the correct technique because it adds a penalty term to the loss function (e.g., L1 or L2 regularization), which constrains the model's weights. This reduces variance and prevents overfitting to noise in the training data, directly addressing the high sensitivity to small input changes (brittleness). By shrinking coefficients, regularization forces the model to learn more general patterns, improving stability and consistency in predictions.

Answer

Increase learning rate

Answer

Feature scaling

Answer

Cross-validation

Question 10

A deep learning model for image classification is overfitting the training data. The team has already tried data augmentation and dropout. Which additional technique should they implement to reduce overfitting?

Accepted Answer

Early stopping. Early stopping (Option D) is the correct additional technique because it halts training when validation performance stops improving, directly preventing the model from memorizing noise in the training data. Since data augmentation and dropout are already in use, early stopping provides a complementary regularization effect by limiting the number of training iterations before overfitting occurs.

Answer

Batch normalization

Answer

Increase number of epochs

Answer

Gradient clipping

Question 11

A company wants to deploy a machine learning model that requires continuous learning as new data arrives. The model must be able to adapt to changing patterns without retraining from scratch. Which approach should be used?

Accepted Answer

Online learning. Online learning (also called incremental learning) updates the model incrementally as each new data point arrives, without requiring full retraining. This makes it ideal for scenarios where data arrives continuously and patterns shift over time, as the model can adapt its parameters on the fly.

Answer

Transfer learning

Answer

Batch learning

Answer

Unsupervised learning

Question 12

A team is building a recommendation system using collaborative filtering. They have a sparse user-item matrix. Which technique should they use to handle the sparsity and improve recommendations?

Accepted Answer

Matrix factorization. Matrix factorization (B) is the correct technique because it decomposes the sparse user-item matrix into lower-dimensional latent factor matrices, effectively capturing underlying patterns and filling in missing entries. This directly addresses sparsity by learning dense representations that generalize beyond observed interactions, which is a core strength in collaborative filtering for recommendation systems.

Answer

Association rule mining

Answer

k-nearest neighbors

Answer

Content-based filtering

Question 13

Which TWO techniques are commonly used to handle missing data in a machine learning dataset? (Choose TWO.)

Accepted Answer

Imputation with mean or median. Imputation with mean or median is a standard technique for handling missing numerical data because it preserves the dataset size and avoids introducing bias from simply discarding rows. By replacing missing values with the central tendency of the observed data, the model can still learn patterns without losing information, though it may reduce variance slightly.

Answer

Normalization

Answer

One-hot encoding

Answer

Dimensionality reduction

Question 14

Which THREE are common activation functions used in neural networks? (Choose THREE.)

Accepted Answer

ReLU. ReLU (Rectified Linear Unit) is a common activation function in neural networks because it introduces non-linearity while being computationally efficient. It outputs the input directly if positive, otherwise zero, which helps mitigate the vanishing gradient problem compared to sigmoid or tanh. This makes it a default choice for hidden layers in many deep learning architectures.

Answer

Softmax

Answer

Linear

Question 15

Which TWO are valid techniques to reduce overfitting in a deep neural network? (Choose TWO.)

Accepted Answer

L2 regularization. L2 regularization (option C) is a valid technique to reduce overfitting by adding a penalty term proportional to the square of the weight magnitudes to the loss function. This discourages the network from learning overly complex patterns, effectively shrinking weights and improving generalization. Dropout (option E) randomly drops a fraction of neurons during training, which prevents co-adaptation of features and forces the network to learn more robust representations, also reducing overfitting.

Answer

Increase batch size

Answer

Increase learning rate

Answer

Gradient clipping