AI RESEARCH

Decentralized EM Algorithm for Gaussian Mixtures under Data Heterogeneity and Partial Labeling

arXiv CS.LG

ArXi:2411.05591v2 Announce Type: replace-cross We systematically study several network-based Expectation-Maximization (EM) algorithms for the Gaussian mixture model within decentralized federated learning (DFL). Our theoretical investigation shows that directly extending the classic EM algorithm to DFL leads to a biased estimator when data are heterogeneously distributed across sites. To address this, we