G-Exponential Families through Probability Coordinates
Abstract
We develop a probability geometric construction of generalized exponential families based on probability coordinate charts. A classical exponential family on the unit interval is transported to value space through a chart, producing what we call a G-exponential family. The construction changes the geometry in which exponential structure is represented while leaving the exponential function itself unchanged. This distinguishes it from the algebraic deformations used in the Tsallis and Kaniadakis frameworks. The transported families inherit the structure of the probability coordinate framework. The probability coordinate of each family member follows exactly the original exponential family on the unit interval. Initial Kolmogorov moments are therefore obtained by pulling classical coordinate moments back to value space. For the canonical family, the probability barycenter is the pullback of the mean parameter, while the Fisher information coincides with the Kolmogorov variance. Tail behaviour is inherited from the chart. Every family member is tail equivalent to the chart density, so heavy tails arise from geometry and remain unchanged in order under probability coordinate tilting. This is the opposite of classical exponential tilting. The sufficient statistic is bounded by construction, maximum likelihood reduces to moment matching in probability coordinates, and the maximum likelihood estimator has a bounded influence function. Thus robustness arises from geometric compactification. We also characterise G-exponential families as minimisers of relative entropy to the chart law under coordinate moment constraints and prove that transport preserves the information geometry of the family. A Cauchy chart example and a Monte Carlo study confirm efficient, calibrated, and bounded influence estimation even though every family member has an infinite mean.
Related
- Equivalence on exponential families as Hessian manifolds and classification of exponential families of order 1 with constant Hessian sectional curvature
- Global Exponential Stabilization of a 3D Nonholonomic Vehicle in Spherical Coordinates
- Approximation of Probability Geometries via Convex Families of Coordinate Charts
- Cramer-Rao Type Bounds for Generalized Exponential Families
- On Natural Exponential Families with Log-Affine Generalized Variance
BibTeX
@misc{cojocea2026gexponential,
title = {{G-Exponential Families through Probability Coordinates}},
author = {Manuela-Simona Cojocea},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2609.03208},
eprint = {2609.03208},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2609.03208},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - G-Exponential Families through Probability Coordinates
AU - Manuela-Simona Cojocea
PY - 2026
DA - 2026/09/02
PB - arXiv
DO - 10.48550/arxiv.2609.03208
UR - https://arxiv.org/abs/2609.03208
AB - We develop a probability geometric construction of generalized exponential families based on probability coordinate charts. A classical exponential family on the unit interval is transported to value space through a chart, producing what we call a G-exponential family. The construction changes the geometry in which exponential structure is represented while leaving the exponential function itself unchanged. This distinguishes it from the algebraic deformations used in the Tsallis and Kaniadakis frameworks. The transported families inherit the structure of the probability coordinate framework. The probability coordinate of each family member follows exactly the original exponential family on the unit interval. Initial Kolmogorov moments are therefore obtained by pulling classical coordinate moments back to value space. For the canonical family, the probability barycenter is the pullback of the mean parameter, while the Fisher information coincides with the Kolmogorov variance. Tail behaviour is inherited from the chart. Every family member is tail equivalent to the chart density, so heavy tails arise from geometry and remain unchanged in order under probability coordinate tilting. This is the opposite of classical exponential tilting. The sufficient statistic is bounded by construction, maximum likelihood reduces to moment matching in probability coordinates, and the maximum likelihood estimator has a bounded influence function. Thus robustness arises from geometric compactification. We also characterise G-exponential families as minimisers of relative entropy to the chart law under coordinate moment constraints and prove that transport preserves the information geometry of the family. A Cauchy chart example and a Monte Carlo study confirm efficient, calibrated, and bounded influence estimation even though every family member has an infinite mean.
ER -