In the case of univariate Gaussian mixtures, unbounded likelihood is an important theoretical and practical problem. Using the weak information that the latent sample size of each component has to be greater than the space dimension, we derive a simple non-asymptotic stochastic lower bound on variances. We prove also that maximizing the likelihood under this data-driven constraint leads to consistent estimates. Résumé Dans le cas des mélanges gaussiens univariés, le fait que la vraisem-blance soit non bornée est un problème théorique et pratique important. Dans ce contexte d’estimation, il est naturel d’imposer que chaque com-posante du mélange a généré un minimum de deux individus. Cette in-formation pour le moins minimale nous permet d’ob...
Although normal mixture models have received great attention and are commonly used in different fiel...
Due to non-regularity of the finite mixture of normal dis-tributions in both mean and variance, the ...
Although normal mixture models have received great attention and are commonly used in different fiel...
In the case of univariate Gaussian mixtures, unbounded likelihood is an important theoretical and pr...
International audienceUnbounded likelihood for multivariate Gaussian mixture is an important theoret...
In the context of the univariate Gaussian mixture with grouped data, it is shown that the global max...
Abstract: A finite mixture of normal distributions in both mean and variance pa-rameters is a typica...
Presented at MaxEnt01. To appear in Bayesian Inference and Maximum Entropy Methods, B. Fry (Ed.), AI...
We consider the problem of identifying the parameters of an unknown mixture of two ar-bitrary d-dime...
International audienceWe consider covariance parameter estimation for a Gaussian process under inequ...
We consider the problem of identifying the parameters of an unknown mixture of two arbi-trary d-dime...
We consider the problem of identifying the parameters of an unknown mixture of two ar-bitrary d-dime...
Mixtures of Gaussian (or normal) distributions arise in a variety of application areas. Many heurist...
Information-theoretic measures, such as the entropy, the cross-entropy and the Kullback–Leibler dive...
Abstract. Slepian and Sudakov-Fernique type inequalities, which com-pare expectations of maxima of G...
Although normal mixture models have received great attention and are commonly used in different fiel...
Due to non-regularity of the finite mixture of normal dis-tributions in both mean and variance, the ...
Although normal mixture models have received great attention and are commonly used in different fiel...
In the case of univariate Gaussian mixtures, unbounded likelihood is an important theoretical and pr...
International audienceUnbounded likelihood for multivariate Gaussian mixture is an important theoret...
In the context of the univariate Gaussian mixture with grouped data, it is shown that the global max...
Abstract: A finite mixture of normal distributions in both mean and variance pa-rameters is a typica...
Presented at MaxEnt01. To appear in Bayesian Inference and Maximum Entropy Methods, B. Fry (Ed.), AI...
We consider the problem of identifying the parameters of an unknown mixture of two ar-bitrary d-dime...
International audienceWe consider covariance parameter estimation for a Gaussian process under inequ...
We consider the problem of identifying the parameters of an unknown mixture of two arbi-trary d-dime...
We consider the problem of identifying the parameters of an unknown mixture of two ar-bitrary d-dime...
Mixtures of Gaussian (or normal) distributions arise in a variety of application areas. Many heurist...
Information-theoretic measures, such as the entropy, the cross-entropy and the Kullback–Leibler dive...
Abstract. Slepian and Sudakov-Fernique type inequalities, which com-pare expectations of maxima of G...
Although normal mixture models have received great attention and are commonly used in different fiel...
Due to non-regularity of the finite mixture of normal dis-tributions in both mean and variance, the ...
Although normal mixture models have received great attention and are commonly used in different fiel...