Institute of Information Theory and Automation

Conditional independence structures: algebraic and geometric methods

Project leader: RNDr. Milan Studený, DrSc.
Department: MTR
Supported by (ID): GA13-20012S
Grantor: Czech Science Foundation
Type of project: theoretical
Duration: 2013 - 2015
Publications at UTIA: list


The application of algebraic and geometric methods is one of the present trends in modern statistics. The aim of the project is to apply the methods of combinatorial optimization to problems with motivation in statistics and artificial intelligence. The goals are divided into three groups: the goals concerning statistical learning Bayesian network structure and conditional independence, the goals concerning exponential families and graphical models, and the goals concerning game theory and artificial intelligence. The project is meant as a natural continuation of both research on conditional independence structures and of collaboration with foreign experts, whose central expertise is in combinatorial optimization.

Project team:
Responsible for information: MTR
Last modification: 18.01.2016
Institute of Information Theory and Automation