Fall 2006

Contents:

 

Dr. Ovidiu Calin Receives NSF Research Grant

Dr. Calin was recently awarded a research grant from the National Science Foundation (NSF). This is a subcontract of a larger grant from the University of Michigan, Ann Arbor with Dr. Jun Zhang as a Principal Investigator. The name of the grant is "Information Geometry with Application to Model Selection", with Ovidiu Calin (Eastern Michigan University) and Hiroshi Matsuzoe (Nagoya Institute of Technology) serving as Co-PIs. The total amount of the grant is $250,000 from which $47,000 is subcontracted by EMU.

This three-year grant will be used to explore informational geometry, a new branch of Mathematics situated at the intersection of statistics and differential geometry.

Informational geometry emerged from investigating geometrical structures of the manifold of probability distributions. In this way, the family of probability distributions, which depend on finitely many parameters, can be viewed as a geometric object called a statistical manifold.

The goal of informational geometry is to geometrically visualize problems from statistics and information theory, and to develop new geometric tools which will be used to extend and advance these fields. Hence a detailed study of statistical manifolds is needed for a good qualitative and quantitative understanding of background statistical problems.

Informational geometry has successful applications to statistical inference problems and brings a new point of view to many fields of Information Theory. For example, in the field of neurocomputing, the set of neural networks form a manifold called a neuromanifold. Recently, informational geometry has become a very useful tool in the investigation of neuromanifolds.

"I hope that this grant will help us carry out enough research so that we can write a new book in the field of informational geometry", says enthusiastic Dr. Calin.