Explore rigidity theory applications in Gaussian graphical models through this 42-minute Fields Institute lecture. Delve into the concept of maximum likelihood threshold for graphs, its significance in genomics, and how tools from rigidity theory provide insights. Learn about the relationship between convex optimization, global and local rigidity, and their implications for understanding maximum likelihood thresholds. Gain valuable knowledge on this collaborative research effort, bridging mathematical concepts with practical applications in data analysis and genomics.
Rigidity Theory for Gaussian Graphical Models - The Maximum Likelihood Threshold of a Graph
Overview
Syllabus
Introduction
Goal
Gaussian graphical model
convex optimization
maximum likelihood threshold
upper bound
global and local rigidity
implications
Global rigidity
Discussion
Conclusion
Taught by
Fields Institute
Related Courses
-
Invariant Theory for Maximum Likelihood Estimation
-
Probabilistic Graphical Models and Maximum Likelihood Estimation - Lecture 13
-
Introduction to Optimization
-
Invariant Theory for Maximum Likelihood Estimation
-
Deep Generative Models - Maximum Likelihood Learning - Lecture 4
-
Algorithmic Thresholds for Spherical Spin Glasses in High-Dimensional Optimization
