Decentralized EM algorithm for Gaussian mixtures under data heterogeneity and partial labeling
Biometrics · 09.04.2026 · 82 (2)
DOI: 10.1093/biomtc/ujag092 ↗Kurz gefasst
Fachartikel in der Zeitschrift Biometrics, online erschienen am 27.05.2026. Verfasst von Xuetong Li und 3 weiteren Autor:innen. Beteiligte Einrichtung: School of Mathematics and Statistics, Xi'an Jiaotong University, 28 Xianning West Road, Xi'an 710049, China. Hauptthemen laut MEDLINE: Algorithms, Models, Statistical. Die Arbeit nennt 26 Literaturangaben. Gefördert von: China Postdoctoral Science Foundation, National Natural Science Foundation of China.
Abstract (Originalsprache)
We systematically study several network-based Expectation–Maximization (EM) algorithms for the Gaussian mixture model within decentralized federated learning (DFL). Our theoretical investigation reveals that directly extending the classic EM algorithm to DFL leads to a seriously biased estimator if the data are heterogeneously distributed across different sites. To address this issue, we introduce a momentum network EM (MNEM) algorithm, which integrates information from both current and historical estimators from previous DFL iterations. We further develop a semi-supervised MNEM (semi-MNEM) algorithm, which utilizes valuable information provided by partially labeled data. Rigorous theoretical analysis demonstrates that the MNEM estimator can achieve the same asymptotic efficiency as the whole sample estimator under appropriate regularity conditions, even if the data are heterogeneously distributed. Moreover, the semi-MNEM estimator significantly improves the convergence speed of the MNEM algorithm, even if different mixture components are poorly separated. Extensive simulations are conducted, and a widely used chest X-ray dataset is analyzed to demonstrate the finite-sample performance of the proposed methods.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- School of Mathematics and Statistics, Xi'an Jiaotong University, 28 Xianning West Road, Xi'an 710049, China.
- School of Statistics and Data Science, Shanghai University of Finance and Economics, 777 Guoding Road, Shanghai 200433, China.
- School of Mathematics and Statistics, Beijing Technology and Business University, 33 Fucheng Road, Beijing 100048, China.
- Guanghua School of Management, Peking University, 5 Yiheyuan Road, Beijing 100871, China.
Förderung
- China Postdoctoral Science Foundation
- National Natural Science Foundation of China
Zitieren
APA
Li, X., Wu, S., Du, B., & Wang, H. (2026). Decentralized EM algorithm for Gaussian mixtures under data heterogeneity and partial labeling. Biometrics, 82(2). https://doi.org/10.1093/biomtc/ujag092
Vancouver
Li X, Wu S, Du B, Wang H. Decentralized EM algorithm for Gaussian mixtures under data heterogeneity and partial labeling. Biometrics. 2026;82(2). doi:10.1093/biomtc/ujag092
BibTeX
@article{Li2026,
author = {Li, Xuetong and Wu, Shuyuan and Du, Bin and Wang, Hansheng},
title = {Decentralized EM algorithm for Gaussian mixtures under data heterogeneity and partial labeling},
journal = {Biometrics},
year = {2026},
volume = {82},
number = {2},
doi = {10.1093/biomtc/ujag092}
}
EUVIMED