PAC-Bayesian aggregation of affine estimators

DOI10.1007/978-3-319-96941-1_9MaRDI QIDQ2417440zbMATH OpenOpenAlexFDO

Authors L. Montuelle, E. Le Pennec

Publication date 12 June 2019

Full work available at URL https://arxiv.org/abs/1410.0661

zbMATH Keywords

regression modeloracle inequality

Mathematics Subject Classification ID

Nonparametric regression and quantile regression (62G08) General nonlinear regression (62J02) Order statistics; empirical distribution functions (62G30)

Abstract: Aggregating estimators using exponential weights depending on their risk appears optimal in expectation but not in probability. We use here a slight overpenalization to obtain oracle inequality in probability for such an explicit aggregation procedure. We focus on the fixed design regression framework and the aggregation of affine estimators and obtain results for a large family of affine estimators under a non necessarily independent sub-Gaussian noise assumptions.

Recommendations

• Sharp oracle inequalities for aggregation of affine estimators
• Aggregation by exponential weighting, sharp PAC-Bayesian bounds and sparsity
• Aggregation of affine estimators
• Optimal bounds for aggregation of affine estimators
• Generalized mirror averaging and D-convex aggregation

Cited in

(10)

• A PAC-Bayes Bound for Tailored Density Estimation
• Aggregation by exponential weighting, sharp PAC-Bayesian bounds and sparsity
• Aggregation of affine estimators
• Fully Bayesian aggregation
• Sharp oracle inequalities for aggregation of affine estimators
• scientific article; zbMATH DE number 7625186 (Why is no real title available?)
• Simple proof of the risk bound for denoising by exponential weights for asymmetric noise distributions
• PAC-Bayesian risk bounds for group-analysis sparse regression by exponential weighting
• Optimal bounds for aggregation of affine estimators
• Aggregated estimators and empirical complexity for least square regression

This page was built for publication: PAC-Bayesian aggregation of affine estimators

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2417440)