STAT 603 - Statistical Computing and Optimization
STAT COMPUTING & OPTIMIZATION
Many modern statistical machine learning problems for Big Data analytics can be formulated by function optimization and linear algebraic computation. This course will provide necessary knowledge of convex optimization and matrix computation, and gain fundamental understandings of important numerical algorithms commonly used in statistical machine learning. We will emphasize on both efficient implementation and understanding for statistical computing problems. The topics to be covered include: fundamental methods for matrix and linear systems computation, matrix decomposition, convex analysis, duality and KKT conditions, 1st/2nd order methods, EM methods. Important statistical computing applications including GLM, SVM, sparsity learning, greedy function approximation, and deep neural networks will be covered.
Repeatable for Credit: N Allowed Units: 3 Multiple Term Enrollment: N Grading Basis: Student Option
PREREQ: STAT 601 and STAT 602 . Basic programming knowledge (such as R, Python, MATLAB, or C/C++) is assumed.