Large deviation entropy defines the geometry of empirical data manifolds, breaking spherical geometry for pairwise statistics, and equates information projections from information geometry with those in Kolmogorov probability theory under i.i.d. and Markov assumptions.
Aspects of convex geometry polyhedra, linear programming, shellings, voronoi diagrams, delaunay triangulations
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.IT 1years
2025 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
The Geometry of Statistical Data and Information: A Large Deviation Perspective
Large deviation entropy defines the geometry of empirical data manifolds, breaking spherical geometry for pairwise statistics, and equates information projections from information geometry with those in Kolmogorov probability theory under i.i.d. and Markov assumptions.