MATH5465

This course gives a basic overview of the mathematical foundations for some commonly used techniques in machine learning and data science. The course will cover basic topics in fully supervised learning (support vector machines, k-nearest neighbor classification), unsupervised learning techniques (principal component analysis, k-means clustering, multi-dimensional scaling), modern methods in deep learning with applications to image classification, an introduction to graph-based learning (spectral clustering, label propagation, PageRank, the discrete Fourier transform, and applications to image processing), and a basic introduction to the convergence theory for gradient-based optimization.
Prerequisites: Linear algebra (for example MATH 2142, 2243 or 2373) and multivariable
calculus (for example MATH 2263 or 2374), or consent of the instructor.