This manuscript focuses on two functional estimation problems. A non asymptotic guarantee of the proposed estimator’s performances is provided for each problem through an oracle inequality.In the conditional density estimation setting, mixtures of Gaussian regressions with exponential weights depending on the covariate are used. Model selection principle through penalized maximum likelihood estimation is applied and a condition on the penalty is derived. If the chosen penalty is proportional to the model dimension, then the condition is satisfied. This procedure is accompanied by an algorithm mixing EM and Newton algorithm, tested on synthetic and real data sets.In the regression with sub-Gaussian noise framework, aggregating linear estimat...
Dans cette thèse, nous considérons le problème de l’estimation paramétrique de la fonction de régres...
This thesis takes place within the theories of nonasymptotic statistics and model selection. Its goa...
International audienceIn this paper, we consider a high-dimensional statistical estimation problem i...
This manuscript focuses on two functional estimation problems. A non asymptotic guarantee of the pro...
Ce manuscrit se concentre sur deux problèmes d'estimation de fonction. Pour chacun, une garantie non...
International audienceAggregating estimators using exponential weights depending on their risk perfo...
In this thesis, we study the approximation capabilities, model estimation and selection properties, ...
International audienceIn the framework of conditional density estimation, we use candidates taking t...
International audienceThis study is devoted to the problem of model selection among a collection of ...
We wish to estimate conditional density using Gaussian Mixture Regression model with logistic weight...
In this technical report, we consider conditional density estimation with a maximum like-lihood appr...
We treat two subjects. The first subject is about statistical learning in high-dimension, that is wh...
In this PhD thesis we present the results we obtained in three linked fields: data compression for i...
International audienceWe consider the problem of combining a (possibly uncountably infinite) set of ...
Dans cette thèse, nous considérons le problème de l’estimation paramétrique de la fonction de régres...
This thesis takes place within the theories of nonasymptotic statistics and model selection. Its goa...
International audienceIn this paper, we consider a high-dimensional statistical estimation problem i...
This manuscript focuses on two functional estimation problems. A non asymptotic guarantee of the pro...
Ce manuscrit se concentre sur deux problèmes d'estimation de fonction. Pour chacun, une garantie non...
International audienceAggregating estimators using exponential weights depending on their risk perfo...
In this thesis, we study the approximation capabilities, model estimation and selection properties, ...
International audienceIn the framework of conditional density estimation, we use candidates taking t...
International audienceThis study is devoted to the problem of model selection among a collection of ...
We wish to estimate conditional density using Gaussian Mixture Regression model with logistic weight...
In this technical report, we consider conditional density estimation with a maximum like-lihood appr...
We treat two subjects. The first subject is about statistical learning in high-dimension, that is wh...
In this PhD thesis we present the results we obtained in three linked fields: data compression for i...
International audienceWe consider the problem of combining a (possibly uncountably infinite) set of ...
Dans cette thèse, nous considérons le problème de l’estimation paramétrique de la fonction de régres...
This thesis takes place within the theories of nonasymptotic statistics and model selection. Its goa...
International audienceIn this paper, we consider a high-dimensional statistical estimation problem i...