Diagnostic Test: Machine Learning

10 multiple-choice questions covering machine learning fundamentals. Select the best answer for each question, then check your score using the answer key below.

Question 1. Which of the following is a supervised learning algorithm?

(A) K-means clustering (B) Principal component analysis (PCA) (C) Linear regression (D) Apriori algorithm

Question 2. The bias-variance tradeoff describes:

(A) The tradeoff between training speed and prediction accuracy (B) The tradeoff between model complexity and generalization error (C) The tradeoff between the number of features and the number of samples (D) The tradeoff between precision and recall

Question 3. Overfitting is characterized by:

(A) High training error and high test error (B) Low training error and low test error (C) Low training error and high test error (D) High training error and low test error

Question 4. What is the purpose of a validation set?

(A) To train the model parameters (B) To tune hyperparameters and estimate generalization performance (C) To deploy the model in production (D) To store raw data before preprocessing

Question 5. In gradient descent, what happens when the learning rate is too large?

(A) The model converges slowly but reliably (B) The optimization may overshoot the minimum and diverge (C) The model always finds the global minimum (D) Training time decreases without affecting convergence

Question 6. L1 regularization (Lasso) has the property of:

(A) Producing dense weight vectors with small non-zero values (B) Producing sparse weight vectors by setting some weights exactly to zero (C) Being computationally more expensive than L2 regularization (D) Only working with linear models

Question 7. In k-fold cross-validation with k equals 10, the model is:

(A) Trained once on 90 percent of the data (B) Trained 10 times, each time on 9 folds and evaluated on 1 fold (C) Trained 10 times on the same training set (D) Evaluated only on the first fold

Question 8. Feature standardization transforms features to have:

(A) A minimum value of 0 and a maximum value of 1 (B) Zero mean and unit variance (C) Only binary values of 0 and 1 (D) The same distribution as the target variable

Question 9. The F1 score is the harmonic mean of:

(A) Accuracy and precision (B) Precision and recall (C) True positive rate and false positive rate (D) Sensitivity and specificity

Question 10. Which technique helps prevent overfitting in neural networks?

(A) Increasing the number of hidden layers (B) Using dropout during training (C) Removing all regularization (D) Training for more epochs without monitoring validation loss

Scoring: Count your correct answers out of 10. A score of 8 or above indicates strong mastery of machine learning fundamentals. Review the explanations in the practice problems for any questions you answered incorrectly.

ML diagnostic assessment tests foundational understanding: Before building complex models, you need to understand the basics, data preprocessing, model evaluation, overfitting, and the bias-variance trade-off.

Why it matters: These fundamentals determine whether your models will actually work in practice, not just in theory.

The key insight: Always evaluate on data the model has never seen, training accuracy tells you almost nothing about real-world performance.

Confusing bias and variance: Bias is error from overly simplistic models (underfitting). Variance is error from models that are too complex (overfitting). Reducing one often increases the other. The bias-variance tradeoff is fundamental.

Using accuracy for imbalanced datasets: If 95% of samples are class A, a model that always predicts A achieves 95% accuracy but is useless. Use precision, recall, F1-score, or AUC-ROC for imbalanced problems.

Forgetting to split data before feature engineering: If you compute statistics on the entire dataset before splitting, information from the test set leaks into training. Always split first, then compute statistics only on the training set.

• Site Home: Main landing page for machine-learning notes.
• Practice: Practice problems for revision.

• Machine Learning Glossary, Key Terms and Definitions
• Machine Learning Fundamentals Study Guide
• Machine Learning