Recursive Aggregation of Estimators by the Mirror Descent Algorithm with Averaging - Université Pierre et Marie Curie Accéder directement au contenu
Article Dans Une Revue Problems of Information Transmission Année : 2005

Recursive Aggregation of Estimators by the Mirror Descent Algorithm with Averaging

Résumé

We consider a recursive algorithm to construct an aggregated estimator from a finite number of base decision rules in the classification problem. The estimator approximately minimizes a convex risk functional under the l1-constraint. It is defined by a stochastic version of the mirror descent algorithm (i.e., of the method which performs gradient descent in the dual space) with an additional averaging. The main result of the paper is an upper bound for the expected accuracy of the proposed estimator. This bound is of the order $\sqrt{(\log M)/t}$ with an explicit and small constant factor, where $M$ is the dimension of the problem and $t$ stands for the sample size. A similar bound is proved for a more general setting that covers, in particular, the regression model with squared loss.
Fichier principal
Vignette du fichier
mda-ai.pdf (269.74 Ko) Télécharger le fichier
Loading...

Dates et versions

hal-00004908 , version 1 (16-05-2005)
hal-00004908 , version 2 (07-03-2006)

Identifiants

Citer

Anatoli B. Juditsky, Alexander Nazin, Alexandre Tsybakov, Nicolas Vayatis. Recursive Aggregation of Estimators by the Mirror Descent Algorithm with Averaging. Problems of Information Transmission, 2005, 41 (4), pp.368-384. ⟨10.1007/s11122-006-0005-2⟩. ⟨hal-00004908v2⟩
429 Consultations
352 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More