Course content
Tentative list of topics
Review of probability theory
Review of linear algebra
Vector spaces, linear independence, linear mappings
Norms, inner products, positive definiteness, orthogonality, projections, and rotations
Matrix decompositions & vector calculus
Determinants, eigenanalysis, EVD and SVD
Gradients of vector-valued functions and matrices
Principal Component Analysis
Numerical optimization
Constrained vs. unconstrained optimization, optimality conditions
Convex sets and functions
Unconstrained optimization algorithms
Gradient descent
Accelerated gradient descent
Newton’s method, gradient projection
Stochastic gradient descent (SGD)
Machine learning motivations, basic formulation, mini-batch SGD
Linear classification
Constrained optimization
Bayesian inference
Approximate inference
Additional topics
Generative processes, and generative modeling
TBD