A non-asymptotic approach for model selection via penalization in high-dimensional mixture of experts models

DOI10.1214/22-EJS2057MaRDI QIDQ2084460

Hien Duy Nguyen, Florence Forbes, TrungTin Nguyen, Faicel Chamroukhi

Publication date: 18 October 2022

Published in: Electronic Journal of Statistics (Search for Journal in Brave)

Full work available at URL: https://arxiv.org/abs/2104.02640

zbMATH Keywords

model selectionclusteringmixture of expertsoracle inequalitygraphical Lassopenalized maximum likelihoodmixture of regressionsblock-diagonal covariance matrixGaussian locally-linear mapping modelslinear cluster-weighted models

Mathematics Subject Classification ID

Estimation in multivariate analysis (62H12) Classification and discrimination; cluster analysis (statistical aspects) (62H30) Approximations to statistical distributions (nonasymptotic) (62E17)

Related Items (1)

Summary statistics and discrepancy measures for approximate Bayesian computation via surrogate posteriors

Uses Software

• R
• mixOmics
• CAPUSHE

Cites Work

• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• High-dimensional regression with Gaussian mixtures and partially-latent response variables
• Local statistical modeling via a cluster-weighted approach with elliptical distributions
• Kullback-Leibler aggregation and misspecified generalized linear models
• Mixture of Gaussian regressions model with logistic weights, a penalized maximum likelihood approach
• Convergence rates of parameter estimation for some weakly identifiable finite mixtures
• \(\ell_{1}\)-penalization for mixture regression models
• Adaptive Dantzig density estimation
• Slope heuristics: overview and implementation
• Concentration inequalities and model selection. Ecole d'Eté de Probabilités de Saint-Flour XXXIII -- 2003.
• Mixtures of regressions with changepoints
• Finite mixture regression: a sparse variable selection by model selection for clustering
• Approximation of conditional densities by smooth mixtures of regressions
• SPADES and mixture models
• On the theory of elliptically contoured distributions
• Estimating the dimension of a model
• Minimum contrast estimators on sieves: Exponential bounds and rates of convergence
• Hierarchical mixtures-of-experts for exponential family regression models: Approximation and maximum likelihood estimation
• Inverse regression approach to robust nonlinear high-to-low dimensional mapping
• Optimal Kullback-Leibler aggregation in mixture density estimation by maximum likelihood
• Rates of convergence for the Gaussian mixture sieve.
• Weak convergence and empirical processes. With applications to statistics
• The Lasso as an \(\ell _{1}\)-ball model selection procedure
• Convergence of latent mixing measures in finite and infinite mixture models
• Approximations of conditional probability density functions in Lebesgue spaces via mixture of experts models
• Minimal penalties for Gaussian model selection
• Optimal exponential bounds for aggregation of estimators for the Kullback-Leibler loss
• Joint rank and variable selection for parsimonious estimation in a high-dimensional finite mixture regression model
• Model-based regression clustering for high-dimensional data: application to functional data
• Anℓ₁-oracle inequality for the Lasso in multivariate finite mixture of multivariate Gaussian regression models
• On Convergence Rates of Mixtures of Polynomial Experts
• A Sparse PLS for Variable Selection when Integrating Omics Data
• POSTERIOR CONSISTENCY IN CONDITIONAL DENSITY ESTIMATION BY COVARIATE DEPENDENT MIXTURES
• Mixture Densities, Maximum Likelihood and the EM Algorithm
• Risk bounds for mixture density estimation
• Sliced Inverse Regression for Dimension Reduction
• Model Selection and Multimodel Inference
• Block-Diagonal Covariance Selection for High-Dimensional Gaussian Graphical Models
• Multivariate extremes, aggregation and dependence in elliptical distributions
• On-line EM Algorithm for the Normalized Gaussian Network
• Multivariate T-Distributions and Their Applications
• A non asymptotic penalized criterion for Gaussian mixture model selection
• Data-driven penalty calibration: A case study for Gaussian mixture model selection
• Approximation by finite mixtures of continuous density functions that vanish at infinity
• Minimal penalties and the slope heuristics: a survey
• ADAPTIVE BAYESIAN ESTIMATION OF CONDITIONAL DENSITIES
• A Universal Approximation Theorem for Mixture-of-Experts Models
• Anℓ₁-oracle inequality for the Lasso in finite mixture Gaussian regression models
• Partition-based conditional density estimation
• Adaptive density estimation for clustering with Gaussian mixtures
• Some Comments on C P
• On the Conditional Distribution of the Multivariate t Distribution
• Maximum Likelihood Estimation of Misspecified Models
• On strong identifiability and convergence rates of parameter estimation in finite mixtures
• A new look at the statistical model identification
• Approximation of probability density functions via location-scale finite mixtures in Lebesgue spaces

This page was built for publication: A non-asymptotic approach for model selection via penalization in high-dimensional mixture of experts models