Quiz: Introduction to Machine Learning Select one answer per question, then click Submit. 1) Which definition best matches machine learning? A) Programming explicit rules for every case B) Giving systems the ability to learn from data without explicit programming C) Compressing data for storage D) Encrypting communication channels ML learns patterns from data rather than fixed rule sets. 2) Which factor did not chiefly drive recent ML growth? A) Data availability B) Increased compute power C) Better algorithms D) Reduced need for data labeling in all cases Data, compute, and algorithms were key; less labeling isn’t a universal driver. 3) Which step adjusts model parameters to reduce error? A) Data collection B) Model selection C) Training D) Deployment Training tunes parameters using optimization like gradient descent. 4) Supervised learning requires: A) Only unlabeled data B) Labeled input–output pairs C) Reward signals only D) No data at all Supervised learning maps inputs to known outputs. 5) An example of unsupervised learning is: A) Logistic regression for spam detection B) K-means for customer segmentation C) Q-learning for a game agent D) Linear regression for price prediction K-means clusters unlabeled data based on similarity. 6) Which statement about reinforcement learning is correct? A) It relies on labeled pairs only B) It ignores rewards C) It learns via trial, error, and rewards D) It cannot be used for control tasks RL optimizes actions based on reward feedback. 7) Which algorithm is a strong margin-based classifier? A) K-Nearest Neighbors B) Support Vector Machine C) Decision Tree D) Naive Bayes SVM finds a maximal margin decision boundary. 8) A key risk when a model memorizes training data too specifically is: A) Underfitting B) Overfitting C) Regularization D) Cross-validation Overfitting harms generalization to new data. 9) Which challenge relates to fairness and responsibility? A) Hyperparameter tuning B) Interpretability C) Ethical bias and societal impact D) GPU memory limits Bias and equity are central ethical considerations in ML. 10) A trend aimed at making models more transparent is known as: A) Edge AI B) Explainable AI (XAI) C) Federated averaging D) Early stopping XAI provides insight into model decisions. Submit Reset