Generic chaining and the _1-penalty

DOI10.1016/J.JSPI.2012.12.004MaRDI QIDQ1948162zbMATH OpenOpenAlexFDO

Authors Sara Van De Geer

Publication date 2 May 2013

Published in Journal of Statistical Planning and Inference (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1205.3703

zbMATH Keywords

\(\ell _{1}\)-regularizationcontraction inequalitydeviation inequalitygeneric chaininghigh-dimensional modelM-estimatorsymmetrization

Mathematics Subject Classification ID

Inequalities; stochastic orderings (60E15) General nonlinear regression (62J02) Ridge regression; shrinkage estimators (Lasso) (62J07)

Abstract: We address the choice of the tuning parameter $lambda$ in $el{l}_1$-penalized M-estimation. Our main concern is models which are highly nonlinear, such as the Gaussian mixture model. The number of parameters $p$ is moreover large, possibly larger than the number of observations $n$. The generic chaining technique of Talagrand[2005] is tailored for this problem. It leads to the choice $lambdaasympsqrtlogp/n$, as in the standard Lasso procedure (which concerns the linear model and least squares loss).

Recommendations

• Generic chaining and the \(\ell _{1}\)-penalty rejoinder
• Comment to ``Generic chaining and the \(\ell _{1}\)-penalty by Sara van de Geer
• Discussion of Sara van de Geer: Generic chaining and the \(L_{1}\) penalty
• Discussion: ``Generic chaining and the \(\ell _{1}\)-penalty by Sara van de Geer
• Tail bounds via generic chaining
• Majorizing measures: The generic chaining
• Combining PAC-Bayesian and generic chaining bounds
• On the general chain pair simplification problem
• Chaining, interpolation, and convexity

Cited in

(4)

• Discussion: ``Generic chaining and the \(\ell _{1}\)-penalty by Sara van de Geer
• Worst possible sub-directions in high-dimensional models
• On non-asymptotic bounds for estimation in generalized linear models with highly correlated design
• A shrinkage likelihood ratio test for high-dimensional subgroup analysis with a logistic-normal mixture model

This page was built for publication: Generic chaining and the \(\ell _{1}\)-penalty

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1948162)