Foundations of Machine Learning

Catalog description

This course provides a rigorous introduction to the principles and methods of machine learning. Students study how models learn from data through the mathematical analysis of decision boundaries, loss functions, optimization strategies, and model complexity. Topics include supervised learning, data and feature representation, basic matrix operations used in model formulation, the geometric interpretation of decision boundaries, loss functions and optimization methods for training models, regularization and model complexity, learned representations, and the theoretical limits of learnability and generalization. Emphasis is placed on developing intuition for how model assumptions, training procedures, and data characteristics influence predictive performance and reliability.