AI0-001 Machine Learning and Deep Learning Practice Test 7 — 15 Questions

Question 1

A company uses a neural network for fraud detection. The dataset has 99% legitimate, 1% fraudulent. The model achieves 99% accuracy but fails to detect most frauds. Which metric should they focus on?

Accepted Answer

Recall. Recall (sensitivity) measures the proportion of actual positives correctly identified. In this fraud detection scenario with 99% legitimate and 1% fraudulent transactions, a 99% accuracy can be achieved by simply predicting all transactions as legitimate, which yields 0% recall for the fraud class. Focusing on recall ensures the model captures the majority of fraudulent cases, addressing the critical failure to detect fraud despite high accuracy.

Answer

Precision

Answer

F1-score

Answer

AUC-ROC

Question 2

A machine learning engineer notices that the gradient values in a deep network are becoming extremely small during backpropagation. What is this problem?

Accepted Answer

Vanishing gradient. The vanishing gradient problem occurs when gradients become extremely small during backpropagation, especially in deep networks with many layers. This causes the weights in earlier layers to update very slowly or not at all, severely hindering training. The correct answer is D because the scenario directly describes the hallmark symptom of vanishing gradients.

Answer

Dead ReLU

Answer

Exploding gradient

Answer

Covariate shift

Question 3

Based on the exhibit, what is the likely problem with the model?

Accepted Answer

Overfitting. The exhibit shows training loss decreasing to near zero while validation loss increases after a certain point, which is a classic sign of overfitting. The model is memorizing the training data rather than learning generalizable patterns, leading to poor performance on unseen data.

Answer

Batch size too small

Answer

Learning rate too high

Answer

Underfitting

Question 4

The exhibit shows a model configuration for a classification task with 10 classes. What is wrong with this setup?

Accepted Answer

The loss function should be categorical crossentropy, not mean squared error. In a multi-class classification task with 10 classes, the correct loss function is categorical crossentropy because it measures the dissimilarity between the true probability distribution and the predicted probability distribution. Mean squared error (MSE) is designed for regression tasks and penalizes errors in a way that is not suitable for classification probabilities, leading to poor gradient behavior and slower convergence.

Answer

The metric should be precision, not accuracy

Answer

The activation should be sigmoid in hidden layers

Answer

The optimizer should be SGD, not Adam

Question 5

A data scientist is training a binary classification model to detect fraudulent transactions. The dataset is highly imbalanced with 99% legitimate and 1% fraudulent. Which evaluation metric should be prioritized to assess model performance?

Accepted Answer

F1-score. In a highly imbalanced dataset (99% legitimate, 1% fraudulent), accuracy is misleading because a model that predicts all transactions as legitimate would achieve 99% accuracy without detecting any fraud. The F1-score combines precision and recall into a single metric, making it the preferred choice for evaluating binary classification performance on imbalanced data, as it penalizes both false positives and false negatives equally.

Answer

Accuracy

Answer

Mean Squared Error

Answer

Log Loss

Question 6

Which TWO of the following are common activation functions used in deep neural networks?

Accepted Answer

ReLU. ReLU (Rectified Linear Unit) is a common activation function in deep neural networks because it introduces non-linearity while being computationally efficient, outputting the input directly if positive and zero otherwise. It helps mitigate the vanishing gradient problem, making it a default choice for hidden layers in many architectures.

Answer

Linear Regression

Answer

Support Vector Machine

Answer

K-means

Question 7

Which THREE of the following are best practices for preventing overfitting in deep learning models?

Accepted Answer

L2 regularization. L2 regularization (also known as weight decay) adds a penalty term proportional to the square of the weight magnitudes to the loss function. This discourages the model from learning overly complex patterns by forcing weights to remain small, which reduces variance and helps prevent overfitting. It is a standard technique in deep learning frameworks like TensorFlow and PyTorch, where it is implemented via the `kernel_regularizer` or `weight_decay` parameter.

Answer

Increasing the number of layers

Answer

Using a larger batch size

Question 8

A hospital wants to deploy a machine learning model to predict patient readmission risk within 30 days. They have a dataset with 10,000 records, 70 features including demographics, lab results, and past admissions. The target variable is binary (readmitted or not). The data scientist trains a logistic regression model and achieves an AUC of 0.85 on the test set. However, the hospital's clinicians require interpretability of predictions to trust the model. Which action should the data scientist take to ensure the model meets the interpretability requirement while maintaining performance?

Accepted Answer

Use the logistic regression model as is, since it is inherently interpretable with coefficients. (PCA + logistic regression) reduces dimensionality but loses interpretability and may degrade performance. Option B (random forest with feature importance) is less interpretable than logistic regression. Option C (deep neural network with LIME/SHAP) adds complexity and may reduce transparency. Option D (keep logistic regression) provides inherent interpretability through coefficients, meeting the requirement without sacrificing performance.

Answer

Reduce the number of features to 10 using PCA and retrain the logistic regression

Answer

Replace logistic regression with a random forest model and use feature importance plots

Answer

Train a deep neural network and apply LIME or SHAP for explanations

Question 9

An e-commerce company uses a gradient boosting model to forecast daily sales. Recently, the model's predictions have become less accurate, showing a significant drop in R-squared on validation data. The data scientist checks for data drift but finds no significant changes in feature distributions. The model was trained on data from the past 24 months and is retrained monthly. Upon inspecting the feature importance, the data scientist notices that the top feature 'promotion_flag' has decreased in importance over time. What is the most likely cause of the performance degradation, and what should be done?

Accepted Answer

Concept drift has occurred; retrain the model more frequently with recent data only, or use an online learning approach. (overfitting to promotions) does not explain the drop over time. Option C (hyperparameter tuning) is unlikely to fix the temporal change. Option D (leakage) would have caused issues from the start. Option B correctly identifies concept drift (changing relationship) and suggests retraining more frequently or using online learning to adapt.

Answer

The model is overfitting to historical promotions; apply more regularization

Answer

The model's hyperparameters need tuning; perform a grid search

Answer

The promotion_flag feature is leaking future information; remove it

Question 10

A financial institution uses a deep learning model for fraud detection. The model is a feedforward neural network with three hidden layers. It was trained on a balanced dataset of 100,000 transactions. During deployment, the model achieves high accuracy on the test set but the fraud detection rate (true positive rate) is only 40% while the false positive rate is 0.1%. The business requires a true positive rate of at least 80%. Which of the following actions is most likely to achieve the required true positive rate while minimizing the increase in false positives?

Accepted Answer

Change the threshold for classifying a transaction as fraud from the default 0.5 to a lower value. (increase hidden layers) may capture more complexity but does not directly increase TPR and could overfit. Option B (SMOTE) rebalances the training set, but the dataset is already balanced, so this is unlikely to improve TPR. Option D (L2 regularization) reduces overfitting but increases bias, which could lower TPR. Option C (change threshold) is the most direct approach: lowering the classification threshold increases the true positive rate, and by tuning, it can achieve 80% TPR with a minimal increase in false positives.

Answer

Increase the number of hidden layers to five to capture more complex patterns

Answer

Use synthetic minority oversampling (SMOTE) to rebalance the training set

Answer

Add L2 regularization to reduce overfitting

Question 11

A healthcare startup is developing a diagnostic system using medical images. The team has collected 10,000 labeled images of skin lesions. They plan to train a convolutional neural network (CNN) from scratch. However, training converges slowly, and the validation accuracy plateaus at 70%. The data scientist suspects overfitting. The dataset contains 8,000 images of benign lesions and 2,000 of malignant. The team has limited GPU resources. Which of the following is the MOST effective course of action to improve validation accuracy? A. Reduce the number of convolutional layers. B. Apply transfer learning using a pre-trained model on ImageNet. C. Increase the learning rate by a factor of 10. D. Add more dropout after every convolutional layer.

Accepted Answer

Apply transfer learning using a pre-trained model on ImageNet.. Transfer learning leverages a model pre-trained on a large dataset (e.g., ImageNet), which provides useful features for medical images and reduces the need for large amounts of data and computational resources. It is particularly effective when the dataset is small and imbalanced. Option A (increasing learning rate) might cause divergence or overshoot minima. Option B (reducing layers) may reduce capacity and underfit. Option C (adding dropout) can help with overfitting but is unlikely to jump from 70% to a significantly higher accuracy given limited data; transfer learning provides a stronger boost.

Answer

Increase the learning rate by a factor of 10.

Answer

Reduce the number of convolutional layers.

Answer

Add more dropout after every convolutional layer.

Question 12

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.

Accepted Answer

Roll back to the previous data pipeline and investigate the root cause of drift.. Since the drift occurred after a pipeline change, rolling back and investigating the root cause is the most prudent first step before making model changes. Retraining on drifted data (A) might incorporate a faulty distribution. Removing drifting features (C) could lose important information and may not fully address the issue. Implementing monitoring (D) is useful for long-term but does not address the immediate degradation.

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 13

A media company uses a natural language processing (NLP) model to classify news articles into topics. The model was trained on articles from 2015-2018. In 2023, the model's F1 score drops significantly. The data scientists find that the word embeddings no longer capture the meaning of some terms (e.g., 'covid', 'metaverse'). The model uses static word embeddings (Word2Vec) trained on the original corpus. Which solution BEST addresses the observed degradation? A. Replace static embeddings with contextual embeddings from a transformer model like BERT, then fine-tune the classifier. B. Retrain the static Word2Vec embeddings on a larger corpus from 2023. C. Apply data augmentation to the original training data by replacing words with synonyms. D. Increase the dimensionality of the static embeddings.

Accepted Answer

Replace static embeddings with contextual embeddings from a transformer model like BERT, then fine-tune the classifier.. Contextual embeddings (e.g., BERT) capture meaning based on context, adapting to new uses of words like 'covid' meaning pandemic. Fine-tuning the classifier on new data would update the model. Option A (retraining static embeddings) might capture new word senses but still assigns a single vector per word, missing context. Option B (increasing dimensionality) does not address the semantic shift. Option D (data augmentation) does not introduce new word meanings.

Answer

Retrain the static Word2Vec embeddings on a larger corpus from 2023.

Answer

Increase the dimensionality of the static embeddings.

Answer

Apply data augmentation to the original training data by replacing words with synonyms.

Question 14

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 15

Based on the exhibit, what is the most likely issue with the trained model?

Accepted Answer

Overfitting because training accuracy is much higher than validation accuracy. The exhibit shows a significant gap between high training accuracy and lower validation accuracy, which is the classic symptom of overfitting. The model has memorized the training data rather than learning generalizable patterns, leading to poor performance on unseen validation data.

Answer

Data leakage artificially inflating training accuracy

Answer

Vanishing gradients causing no learning

Answer

Underfitting due to insufficient epochs