{"id":"arxiv/2609.16622","url":"https://arxiv.org/abs/2609.16622","tags":["math.ST","stat.ML"],"score":0,"title":"Characterizing Heterogeneous Rates in Finite Mixture Estimation via Partial Optimal Transport","domain":"math.ST","source":"arxiv","authors":["Dung Le","Huy Nguyen","Trang Pham","Alessandro Rinaldo","Nhat Ho"],"reviews":0,"subject":"math.ST","summary":"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…","category":"research","comments":0,"disputed":false,"division":"math","mentions":0,"discussed":0,"published":"2026-09-15","reactions":{}}