A local-to-global principle for f-divergence homogenization
Abstract
We study the effect of replacing a family of independent statistical experiments by independent repetitions of their average. We give a simple, complete characterization of the f-divergences for which this operation increases divergence by at most a constant depending only on the generator, uniformly in the number of observations and the underlying measurable space. The characterization is governed by a local-to-global principle: a comparison restricted to two observations on a binary alphabet already determines the validity of the general inequality. It admits an explicit scalar formulation, which simplifies further for symmetric divergences. The result applies to arbitrary convex generators and requires no differentiability or symmetry assumptions. It recovers the homogenization inequality for total variation and identifies exactly the admissible members of the power-divergence family. In particular, whenever uniform homogenization fails, the obstruction is already present in a family of binary experiments with two observations. Moreover, the global comparison constant is controlled uniformly over generators with a common bound in the local inequality.
Related
- Global Hölder estimates for linearized Monge-Ampère equations in divergence form with applications to dual semigeostrophic equations and periodic homogenization
- A Spectral Local-to-Global Principle for Spin Systems on Graphs with Girth At Least Five
- Homogenization effects on non-local functionals
- Emergence of a convex strange term via homogenization of non-local energies at the critical exponent
- Shifted Composition III: Local Error Framework for KL Divergence
BibTeX
@misc{kontorovich2026localtoglobal,
title = {{A local-to-global principle for $f$-divergence homogenization}},
author = {Aryeh Kontorovich},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.31164},
eprint = {2610.31164},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.31164},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - A local-to-global principle for f-divergence homogenization
AU - Aryeh Kontorovich
PY - 2026
DA - 2026/10/22
PB - arXiv
DO - 10.48550/arxiv.2610.31164
UR - https://arxiv.org/abs/2610.31164
AB - We study the effect of replacing a family of independent statistical experiments by independent repetitions of their average. We give a simple, complete characterization of the f-divergences for which this operation increases divergence by at most a constant depending only on the generator, uniformly in the number of observations and the underlying measurable space. The characterization is governed by a local-to-global principle: a comparison restricted to two observations on a binary alphabet already determines the validity of the general inequality. It admits an explicit scalar formulation, which simplifies further for symmetric divergences. The result applies to arbitrary convex generators and requires no differentiability or symmetry assumptions. It recovers the homogenization inequality for total variation and identifies exactly the admissible members of the power-divergence family. In particular, whenever uniform homogenization fails, the obstruction is already present in a family of binary experiments with two observations. Moreover, the global comparison constant is controlled uniformly over generators with a common bound in the local inequality.
ER -