Assess graduate students' understanding of the core concepts and distinctions among supervised, unsupervised, and reinforcement learning, including recognition of typical algorithms (e.g., linear/logistic regression, decision trees/SVM, k-means/PCA, Q-learning/policy gradients) and matching each paradigm to representative example applications (classification/regression, clustering/dimensionality reduction, control/gameplay/robotics); focus on basic definitions, when to apply each paradigm, and simple algorithm-to-application mapping.