Math for Machine Learning

Catalog description

Numerical methods underlying a variety of modern machine learning and data analysis techniques are discussed. Topics include block-matrix multiplication and its role in high-performance computing, Identification of fundamental low-dimensional structure or parameters in high-dimensional problems, and model reduction through Principal Component Analysis, Data fitting and classification through multi-objective and constrained least squares, Introduction to data fitting and classification through nonlinear least squares, Bias-Variance tradeoff in Lasso (L1) and Ridge (L2) regression in data fitting problems, and ill conditioning.