The dimensionality of current applications increases which makes the density estimation a difficult task. Indeed, the needed number of parameters to make estimation grows exponentially with respect to the dimension of the problem. Probabilistic graphical models can be used to solve this problem by providing a factorization of the joint distribution, but they suffer from a problem of scalability. The problem of high dimensional spaces is accentuated by the number of observations used to perform density estimation witch is not increased in the same proportions, and even remains extremely law in some applications. Factorization of the joint distribution is not sufficient to perform good density estimation with sparse data. The Perturb and Comb...
Nous considérons le problème d’estimation de densités conditionnelles en modérément grandes dim...
Computing and storing probabilities is a hard problem as soon as one has to deal with complex distri...
Dans cet article, nous comparons l’introduction d’heuristiques faibles (bootstrap, de complexité q...
Dans les applications actuelles, le nombre de variables continue d'augmenter, ce qui rend difficile ...
Les modèles graphiques probabilistes codent les dépendances entre les variables aléatoires et l’esti...
International audienceIn this work we explore the Perturb and Combine idea celebrated in supervised ...
International audienceIn this work we explore the Perturb and Combine idea, celebrated in supervised...
L’entropie d’une distribution sur un ensemble de variables aléatoires discrètes est toujours bornée ...
Ensembles of weakly fitted randomized models have been studied intensively and used successfully in ...
The goal of this thesis is to study and develop methods for density estimation and curve classificat...
peer reviewedWe consider randomization schemes of the Chow-Liu algorithm from weak (bagging, of quad...
Statistical machine learning is a general framework to study predictive problems, where one aims to ...
The data preparation step of the data mining process represents 80% of the problem and is both time ...
Datasets come in a variety of forms and from a broad range of different applications. Typically, the...
We consider the problem of conditional density estimation in moderately large dimen- sions. Much mor...
Nous considérons le problème d’estimation de densités conditionnelles en modérément grandes dim...
Computing and storing probabilities is a hard problem as soon as one has to deal with complex distri...
Dans cet article, nous comparons l’introduction d’heuristiques faibles (bootstrap, de complexité q...
Dans les applications actuelles, le nombre de variables continue d'augmenter, ce qui rend difficile ...
Les modèles graphiques probabilistes codent les dépendances entre les variables aléatoires et l’esti...
International audienceIn this work we explore the Perturb and Combine idea celebrated in supervised ...
International audienceIn this work we explore the Perturb and Combine idea, celebrated in supervised...
L’entropie d’une distribution sur un ensemble de variables aléatoires discrètes est toujours bornée ...
Ensembles of weakly fitted randomized models have been studied intensively and used successfully in ...
The goal of this thesis is to study and develop methods for density estimation and curve classificat...
peer reviewedWe consider randomization schemes of the Chow-Liu algorithm from weak (bagging, of quad...
Statistical machine learning is a general framework to study predictive problems, where one aims to ...
The data preparation step of the data mining process represents 80% of the problem and is both time ...
Datasets come in a variety of forms and from a broad range of different applications. Typically, the...
We consider the problem of conditional density estimation in moderately large dimen- sions. Much mor...
Nous considérons le problème d’estimation de densités conditionnelles en modérément grandes dim...
Computing and storing probabilities is a hard problem as soon as one has to deal with complex distri...
Dans cet article, nous comparons l’introduction d’heuristiques faibles (bootstrap, de complexité q...