ECE 403/503 Optimization for Machine Learning

The optimization techniques to be studied in the course include gradient descent, Newton, conjugate gradient, stochastic gradient, mini-batch, quasi-Newon, and memoryless quasi-Newton methods. The course also addresses applications of optimization methods in logstic regression, multi-category classification, model selection and validation, K-means clustering, autoencoder, and principal component analysis. The course includes laboratory sessions to implement these algorithms and apply them to several machine learning problems involving real-world datasets.

Return to home page