Characterizing Heterogeneous Rates in Finite Mixture Estimation via Partial Optimal Transport

Abstract

Parameter estimation in finite mixture models can exhibit highly heterogeneous convergence behavior: locally isolated components may be estimated substantially faster than groups of competing components. Existing analyses based on Wasserstein distances typically characterize only the worst-case rate and therefore do not fully capture this local heterogeneity. In this paper, we introduce a Voronoi-based partial optimal transport (VPOT) framework for obtaining refined local and global convergence guarantees for the maximum likelihood estimator of the mixing measure. The key geometric idea is to localize the comparison of two mixing measures to extended Voronoi neighborhoods and use partial optimal transport to accommodate the unequal masses of their local restrictions. Within each neighborhood, the first-order POT discrepancy is raised to a power determined by the number of locally competing atoms, allowing the resulting loss to adapt to the local degree of singularity. Under suitable regularity and strong identifiability conditions, we establish uniform local and global upper bounds for a maximum likelihood estimator under the VPOT loss. These bounds reveal a configuration-dependent form of parameter estimation: less singular local configurations admit faster convergence, whereas the most singular configuration recovers the classical worst-case behavior characterized by Wasserstein-based analyses. We further establish a minimax lower bound showing that the convergence rate for estimating the mixing measure under the VPOT loss is optimal. Our results hold in arbitrary fixed dimension without requiring mixing proportions to be uniformly bounded away from zero or prior knowledge of the true number of mixture components. Overall, VPOT provides a configuration-adaptive framework for capturing heterogeneous parameter-estimation behavior in finite mixture models.

Byzantine-tolerant distributed learning of finite mixture models under partial corruptions
Yimei Zhang, Jiahua Chen, Xiaozhou Wang +2
arXiv10 Sep 2026PreprintarXiv licenceFull textStatisticsarXiv
Sandwich Variance Estimation in Scale Mixture of Skew-Normal Linear Mixed Models.
Keyliane Travassos, Larissa A Matos, Fernanda L Schumacher
Statistics in medicine1 Oct 2026Peer-reviewedCC BYFull textStatisticsEurope PMC
Brown's loss function and estimation of the scale parameter and the entropy in scale mixture models with unknown location
Stavros Kourouklis
arXiv9 Oct 2026PreprintarXiv licenceFull textStatisticsarXiv
Partial Localization and Multifractality in Heterogeneous Population Growth with Resetting
Maximilien Bernard, Giuseppe Del Vecchio Del Vecchio, Jean-Philippe Bouchaud +1
arXiv15 Oct 2026PreprintCC BY 4.0Full textEvolutionary biology ProbabilityarXiv
Computer code validation via mixture model estimation
Negar Soleimani, Pierre Barbillon, Kaniav Kamary +1
arXiv23 Sep 2026PreprintarXiv licenceFull textStatisticsarXiv

Cite

BibTeX RIS

BibTeX

@misc{le2026characterizing,
  title = {{Characterizing Heterogeneous Rates in Finite Mixture Estimation via Partial Optimal Transport}},
  author = {Dung Le and Huy Nguyen and Trang Pham and Alessandro Rinaldo and Nhat Ho},
  howpublished = {arXiv},
  year = {2026},
  month = sep,
  doi = {10.48550/arxiv.2609.16622},
  eprint = {2609.16622},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2609.16622},
}

RIS

TY  - UNPB
TI  - Characterizing Heterogeneous Rates in Finite Mixture Estimation via Partial Optimal Transport
AU  - Dung Le
AU  - Huy Nguyen
AU  - Trang Pham
AU  - Alessandro Rinaldo
AU  - Nhat Ho
PY  - 2026
DA  - 2026/09/15
PB  - arXiv
DO  - 10.48550/arxiv.2609.16622
UR  - https://arxiv.org/abs/2609.16622
AB  - Parameter estimation in finite mixture models can exhibit highly heterogeneous convergence behavior: locally isolated components may be estimated substantially faster than groups of competing components. Existing analyses based on Wasserstein distances typically characterize only the worst-case rate and therefore do not fully capture this local heterogeneity. In this paper, we introduce a Voronoi-based partial optimal transport (VPOT) framework for obtaining refined local and global convergence guarantees for the maximum likelihood estimator of the mixing measure. The key geometric idea is to localize the comparison of two mixing measures to extended Voronoi neighborhoods and use partial optimal transport to accommodate the unequal masses of their local restrictions. Within each neighborhood, the first-order POT discrepancy is raised to a power determined by the number of locally competing atoms, allowing the resulting loss to adapt to the local degree of singularity. Under suitable regularity and strong identifiability conditions, we establish uniform local and global upper bounds for a maximum likelihood estimator under the VPOT loss. These bounds reveal a configuration-dependent form of parameter estimation: less singular local configurations admit faster convergence, whereas the most singular configuration recovers the classical worst-case behavior characterized by Wasserstein-based analyses. We further establish a minimax lower bound showing that the convergence rate for estimating the mixing measure under the VPOT loss is optimal. Our results hold in arbitrary fixed dimension without requiring mixing proportions to be uniformly bounded away from zero or prior knowledge of the true number of mixture components. Overall, VPOT provides a configuration-adaptive framework for capturing heterogeneous parameter-estimation behavior in finite mixture models.
ER  -

Abstract

Related

Cite