Library

Fast computation and theoretical guarantees for the NPMLE in exponential family mixtures

This work makes two advances in the study of the (approximate) nonparametric maximum likelihood estimator (NPMLE) for exponential family mixture models. First, we develop a data-compression strategy that reduces the cost of repeated likelihood evaluations in NPMLE computation to polylogarithmic order in the sample size. Second, we show that, for a broad class of approximate NPMLEs, the resulting marginal density estimator attains an almost parametric rate of convergence.

Bayesian Methods and Mixture ModelsMachine Learning and AlgorithmsStatistical Methods and Inference
Source

Paper

Similar papers

Adaptivity of the NPMLE to finitely discrete mixing distributions in Gaussian/Poisson mixturesYan Zhang, Stanislav Volgushev2026 · arXiv (Cornell University) In Library
On Efficient and Scalable Computation of the Nonparametric Maximum Likelihood Estimator in Mixture ModelsYangjing Zhang, Ying Cui, Bodhisattva Sen et al.2022 · arXiv (Cornell University) · 9 citations In Library
Self-regularizing Property of Nonparametric Maximum Likelihood Estimator in Mixture ModelsYury Polyanskiy, Yihong Wu2020 · arXiv (Cornell University) In Library
Parametric convergence rate of some nonparametric estimators in mixtures of power series distributionsBalabdaoui, Fadoua, Besdziek, Harald, Wang, Yong2025 · Repository for Publications and Research Data (ETH Zurich) In Library
Parametric convergence rate of some nonparametric estimators in mixtures of power series distributionsFadoua Balabdaoui, Harald Besdziek, Yong Wang2025 · arXiv (Cornell University) In Library
On the nonparametric maximum likelihood estimator for Gaussian location mixture densities with application to Gaussian denoisingSujayam Saha, Adityanand Guntuboyina2020 · The Annals of Statistics In Library
Multivariate, heteroscedastic empirical Bayes via nonparametric maximum likelihoodJake A. Soloff, Adityanand Guntuboyina, Bodhisattva Sen2024 · Journal of the Royal Statistical Society Series B (Statistical Methodology) In Library
A Nonparametric Maximum Likelihood Approach to Mixture of RegressionHansheng Jiang, Adityanand Guntuboyina2021 · arXiv (Cornell University) In Library
Rates of Convergence of Maximum Smoothed Log-Likelihood Estimators for Semi-Parametric Multivariate MixturesMarie Du Roy de Chaumaray, Michael Levine, Matthieu Marbac2025 · arXiv (Cornell University) In Library
Rates of Convergence of Maximum Smoothed Log-Likelihood Estimators for Semi-Parametric Multivariate MixturesDu Roy de Chaumaray, Marie, Levine, Michael, Marbac, Matthieu2025 · HAL (Le Centre pour la Communication Scientifique Directe) In Library
© 2026 NYSGPT2525 LLC