Title :
From differential to information geometry
Author_Institution :
EADS Deutschland GmbH, Ulm, Germany
Abstract :
Information geometry is a new and increasing topic between statistics, estimation and differential geometry. Many amazing relationships between these domains were established through the last years. Unfortunately, it is not easy to find an easy approach to information geometry, which requires a deep understanding of differential geometry and statistics. The paper presents an easy readable introduction to information geometry, adapted from the analogies of surfaces embedded in the three dimensional space. It is shown, how well known structures like hyperbolic geometry and affine spaces occurs again in information geometry. Finally, some applications are given.
Keywords :
differential geometry; statistics; afflne spaces; differential geometry; estimation; hyperbolic geometry; information geometry; statistics; Gaussian distribution; Information geometry; Manifolds; Measurement; Vectors;
Conference_Titel :
Cognitive Information Processing (CIP), 2010 2nd International Workshop on
Conference_Location :
Elba
Print_ISBN :
978-1-4244-6457-9
DOI :
10.1109/CIP.2010.5604248