Return to Colloquia & Seminar listing
Maximum Likelihood for Matrices with Rank Constraints
Algebra & Discrete Mathematics| Speaker: | Bernd Sturmfels, Dept. of Mathematics, UC Berkeley |
| Location: | 1147 MSB |
| Start time: | Thu, Feb 14 2013, 3:10PM |
Description
Maximum likelihood estimation is a fundamental
computational task in statistics. We discuss this problem for
manifolds of low rank matrices. These represent mixtures of
independent distributions of two discrete random variables.
This non-convex optimization leads to some beautiful geometry,
topology, and combinatorics. We explain how numerical algebraic
geometry is used to find the global maximum of the likelihood
function. This is joint work with Jon Hauenstein and Jose Rodriguez.