2003 | OriginalPaper | Chapter
Geometric Ideas in Minimum Cross-Entropy
Author : L. Lore Campbell
Published in: Entropy Measures, Maximum Entropy Principle and Emerging Applications
Publisher: Springer Berlin Heidelberg
Included in: Professional Book Archive
Activate our intelligent search to find suitable subject content or patents.
Select sections of text to find matching patents with Artificial Intelligence. powered by
Select sections of text to find additional relevant content using AI-assisted search. powered by
This article reviews three geometric approaches to the understanding of the minimum cross-entropy method for estimating a probability distribution. The first approach is to regard the method as a projection based on an analogue of Pythagoras’ Theorem. The second is to regard the set of probability distributions as a differentiable manifold and to introduce a Riemannian geometry on this manifold. The third uses the idea of Hausdorff dimension to support the use of the method.