MATRIX METHODS IN MACHINE LEARNING

Prerequisites & course information

Requisites: ( MATH 234 , 320 , 340 , 341 , or 375 ) and ( E C E 203 , COMP SCI 200 , 220 , 300 , 301, 310 , 320 , or placement into COMP SCI 300 ), graduate/professional standing, or declared in Capstone Certificate in Computer Sciences for Professionals Course Designation: Breadth - Physical Science Level - Advanced L&S Credit - L&S Liberal Arts and Science Grad 50% - Counts toward 50% graduate coursework requirement Repeatable for Credit: No Last Taught: Fall 2026 Learning Outcomes: 1. Use matrices and vectors to formulate classification, prediction and matrix completion problems using techniques such as least squares, regularized least squares, the singular value decomposition, subspace methods, support vector machines, neural networks and kernel methods. Audience: Both Grad & Undergrad 2. Implement machine learning techniques for classification, prediction and matrix completion problems in software, and validate their performance on datasets using cross validation. Audience: Both Grad & Undergrad 3. Apply advanced techniques to formulate and prove optimality of various matrix based techniques in machine learning. Audience: Graduate Learning Outcomes: 1. Use matrices and vectors to formulate classification, prediction and matrix completion problems using techniques such as least squares, regularized least squares, the singular value decomposition, subspace methods, support vector machines, neural networks and kernel methods. Audience: Both Grad & Undergrad 2. Implement machine learning techniques for classification, prediction and matrix completion problems in software, and validate their performance on datasets using cross validation. Audience: Both Grad & Undergrad 3. Apply advanced techniques to formulate and prove optimality of various matrix based techniques in machine learning. Audience: Graduate