Second order concentration via logarithmic Sobolev inequalities

From MaRDI portal
Publication:2278656



Abstract: We show sharpened forms of the concentration of measure phenomenon centered at first order stochastic expansions. The bound are based on second order difference operators and second order derivatives. Applications to functions on the discrete cube and stochastic Hoeffding type expansions in mathematical statistics are studied as well as linear eigenvalue statistics in random matrix theory.


The authors derive second-order concentration-of-measure results, that is, employing second-order difference (or differential) operators, and where fluctuations of random variables of the form \(f-\mathbb{E}f-f_1\) are considered, where \(f_1\) is the first-order term in the Hoeffding decomposition of \(f\). Both the differentiable and non-differentiable settings are studied. Proofs are based on modified logarithmic Sobolev inequalities and exponential inequalities. Applications are given to functions of independent random variables and in random matrix theory.



Cites work









This page was built for publication: Second order concentration via logarithmic Sobolev inequalities

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2278656)