MAT 5003 Numerical Methods

Algorithms in machine learning and neural networks are built upon a strong foundation of linear algebra. For example, modern recommendation systems may have sparse matrices with millions of users and millions of items; matrix factorization methods make the underlying calculations tractable say this course builds a foundation of linear algebra concepts such as matrices, determinants, vectors and eigen values. Then it deepens it into data science applications around network analysis and logistic algorithms. In addition, some multi-variate calculus and graph theory topics are covered. Open to graduate students. Undergraduate students may as a Pathways student through the dual degree BA/MA program or take it as undergraduate credit only by filling out the "Request To Take Graduate Course for Undergraduate Credit" form available on the Registrar's Office website. Crosslisted with AIM 5003. Previously offered as MAT 5400.

Credits

3