Devika Subramanian

This course offers a comprehensive introduction to the foundational principles and practical techniques of machine learning. Students will explore key concepts such as learning from data, concept classes or models, learning objectives, loss functions, and the formulation of machine learning as an optimization problem. The course will also include ethical considerations in AI and discuss topics including bias and fairness. They will gain an understanding of core ideas in generalization, model evaluation, and trade-offs between performance and resource efficiency. The course emphasizes hands-on experience in implementing and understanding ML algorithms, with a strong focus on thorough evaluation and monitoring. Topics include supervised and unsupervised learning, linear models for regression and classification, non-linear models such as decision trees, ensemble methods, and neural networks. Additional topics include nearest neighbor search and probabilistic modeling techniques such as naïve Bayes. Students will also learn basic unsupervised learning methods including clustering, PCA, and other dimensionality reduction techniques. The course covers machine learning evaluation strategies, including accuracy metrics, efficiency trade-offs, and best practices for model monitoring. Mutually Exclusive: Cannot register for COMP 345 if student has credit for COMP 447/COMP 546.

Theoretical and experimental investigations under staff direction. Repeatable for Credit.

Repeatable for Credit.

Repeatable for Credit.