Algebra in Statistics and Computation Seminar

Information

Organizer(s)

Jose Israel Rodriguez, University of Wisconsin --- Madison.

Usual meeting times and calendar

We usually meet on a Wednesday afternoon for about 40 minutes. See the schedule below for details.

Seminar Abstract (Fall 2020)

Algebraic statistics studies statistical models through the lens of algebra, geometry, and combinatorics. From model selection to inference, this interdisciplinary field has seen applications in a wide range of statistical procedure and motivated many questions in algebraic geometry. This seminar series is open to topics on new developments in algebraic statistics and related areas.

This seminar series is discussion based. Each seminar will have a lead participant who selects a topic and video lecture to discuss. Other participants are expected to watch the video lecture and/or read the paper ahead of time.

Virtual Location and Calendar:

The seminar will be held in a recurring Zoom room.

We use the ASCplus calendar for announcements. In addition to listing the topic of the seminar, it lists upcoming conferences, related seminars, and important deadlines.

Schedule

Wed, 02/Dec/2020: 1:45pm - 2:25pm.
Lead participant: Zinan Wang

This seminar will discuss a themes on (Gram) spectrahedra: symmetry adapted spectrahedra. This is the topic Serkan Hosten didn't get to in the Algebraic Statistics Online Seminar (ASOS) on October 9th at 2pm with a recording saved here.

History

Wed, 30/Sep/2020: 1:45pm - 2:25pm.
Lead participant: Jose Israel Rodriguez

This seminar will discuss logarithmic voronoi cells, from a perspective by Yulia Alexandr presented in the NASO seminar. I want to know more about the relation to ML degree 1 models and to understand some of the open questions that they pose in the paper.

Wed, 21/Oct/2020: 1:45pm - 2:25pm.
Lead participant: Julia Lindberg

This seminar will discuss two themes on (Gram) spectrahedra: central curves and symmetry, from a perspective by Serkan Hosten presented in the Algebraic Statistics Online Seminar (ASOS) on October 9th at 2pm with a recording to be saved here.

Wed, 28/Oct/2020: 1:45pm - 2:25pm.
Lead participant: Ben Hollering (NCSU)

Maximum likelihood estimation is a fundamental computational problem in statistics. Prior to this discussion some code to compute maximum likelihood estimates for Gaussian graphical models will be distributed which we will review together. We will begin by discussing the picture on page 10 and page 4.

Wed, 11/Nov/2020: 1:45pm - 2:25pm.
Lead participant: Xinyang Hu

This seminar will discuss quasi-independence models with rational MLE, from a perspective by Jane Ivy Coons and Seth Sullivant. This is also related to work presented by Bernd Sturmfels Statistical Models with Rational Maximum Likelihood Estimator.