WorldCat Identities

Birgé, Lucien

Overview
Works: 9 works in 11 publications in 2 languages and 22 library holdings
Roles: Author, Opponent, Thesis advisor
Classifications: QA267.8,
Publication Timeline
.
Most widely held works by Lucien Birgé
Approximation dans les spaces métriques et théorie de l'estimation inégalités de Cràmer-Chernoff et théorie asymptotique des tests by Lucien Birgé( Book )

3 editions published in 1980 in French and held by 7 WorldCat member libraries worldwide

Vitesses maximales de décroissance des erreurs et tests optimaux associes. Sur un théorème de minimax et son application aux tests. Approximation dans les espaces métriques et théorie de l'estimation
Quelques problèmes de vitesse de convergence pour des processus empiriques by Pascal Massart( Book )

1 edition published in 1987 in French and held by 3 WorldCat member libraries worldwide

LE THEOREME DE DONSKER ASSURE QUE LE PONT BROWNIEN EMPIRIQUE ASSOCIE A N OBSERVATIONS A VALEURS REELLES, INDEPENDANTES ET DE MEME LOI, CONVERGE UNIFORMEMENT SUR LA CLASSE DES INTERVALLES VERS UN PONT BROWNIEN. LE THEOREME HONGROIS (1975) PRECISE LA VITESSE EXACTE DE CETTE CONVERGENCE EN TERMES D'APPROXIMATIONS FORTES. DANS LA VOIE OUVERTE DE R.M. DUDLEY (1978), NOUS EVALUONS LES VITESSES DE CONVERGENCE LORSQUE LA CLASSE DES INTERVALLES EST REMPLACEE PAR UNE CLASSE DE FONCTIONS SATISFAISANT A UNE CERTAINE CONDITION D'ENTROPIE METRIQUE (L'HYPOTHESE D'INDEPENDANCE PEUT ETRE REMPLACEE PAR UNE CONDITION DE FAIBLE DEPENDANCE DE TYPE MELANGE). PAR AILLEURS, NOUS CALCULONS (A UNE PUISSANCE DE LOG(N) PRES), L'ORDRE DE GRANDEUR PRESQUE SUR DE LA DISTANCE DE PROHOROV ENTRE LA DISTRIBUTION UNIFORME SUR UN CUBE ET SA LOI EMPIRIQUE (PROBLEME OUVERT DEPUIS 1968). NOUS DEVELOPPONS A CETTE OCCASION UNE METHODE D'APPROXIMATION QUI, INVESTIE DANS LE PROBLEME DE CALCUL DE LA VITESSE DE CONVERGENCE DU PONT BROWNIEN EMPIRIQUE ASSOCIE A DES OBSERVATIONS A VALEURS DANS UN ESPACE DE DIMENSION D, VERS UN PROCESSUS GAUSSIEN, PERMET DE DEMONTRER UN ANALOGUE DU THEOREME HONGROIS POUR DES CLASSES DE PARTIES CONVENABLES (TYPIQUEMENT, POUR LA CLASSE DES BOULES EUCLIDIENNES OU DES DEMI-ESPACES, LES VITESSES OBTENUES SONT OPTIMALES, A UNE PUISSANCE DE LOG(N) PRES)
On the risk of histograms for estimating decreasing densities by Lucien Birgé( Book )

1 edition published in 1984 in English and held by 3 WorldCat member libraries worldwide

Estimating a density under order restrictions : non-asymptotic minimax risk by Lucien Birgé( Book )

1 edition published in 1984 in English and held by 2 WorldCat member libraries worldwide

Rates of convergence for minimum contrast estimators by Lucien Birgé( Book )

1 edition published in 1991 in French and held by 2 WorldCat member libraries worldwide

On estimating a density using Hellinger distance and some other strange facts by Lucien Birgé( Book )

1 edition published in 1983 in English and held by 2 WorldCat member libraries worldwide

Sur un théorème de minimax et son application aux tests by Lucien Birgé( Book )

1 edition published in 1979 in English and held by 1 WorldCat member library worldwide

Gaussian model selection by Lucien Birgé( Book )

1 edition published in 2000 in English and held by 1 WorldCat member library worldwide

Validation croisée et pénalisation pour l'estimation de densité by Nelo Magalhães( )

1 edition published in 2015 in French and held by 1 WorldCat member library worldwide

This thesis takes place in the density estimation setting from a nonparametric and nonasymptotic point of view. It concerns the statistical algorithm selection problem which generalizes, among others, the problem of model and bandwidth selection. We study classical procedures, such as penalization or resampling procedures (in particular V-fold cross-validation), which evaluate an algorithm by estimating its risk. We provide, thanks to concentration inequalities, an optimal penalty for selecting a linear estimator and we prove oracle inequalities and adaptative properties for resampling procedures. Moreover, new resampling procedure, based on estimator comparison by the mean of robust tests, is introduced as an alternative to procedures relying on the unbiased risk estimation principle. A second goal of this work is to compare these procedures from a theoretical point of view and to understand the role of V for V-fold penalization. We validate these theoretical results on empirical studies
 
Audience Level
0
Audience Level
1
  Kids General Special  
Audience level: 0.89 (from 0.82 for Estimating ... to 1.00 for Validation ...)

Alternative Names
Lucien Birg fransk matematikar

Lucien Birg fransk matematiker

Lucien Birg franzsischer Mathematiker

Languages