On McDiarmid's concentration inequality

In this paper we improve the rate function in the McDiarmid concentration inequality  for separately Lipschitz functions of independent random variables. In particular the rate function tends to infinity at the boundary. We also prove that in some cases the usual normalization factor is not adequate and may be improved.

[1]  Igor Vajda,et al.  Note on discrimination information and variation (Corresp.) , 1970, IEEE Trans. Inf. Theory.

[2]  I. Pinelis Binomial upper bounds on generalized moments and tail probabilities of (super)martingales with differences bounded from above , 2005, math/0512301.

[3]  Luc Devroye,et al.  Combinatorial methods in density estimation , 2001, Springer series in statistics.

[4]  V. Bentkus On measure concentration for separately Lipschitz functions in product spaces , 2007 .

[5]  J. Kurchan,et al.  In and out of equilibrium , 2005, Nature.

[6]  M. Habib Probabilistic methods for algorithmic discrete mathematics , 1998 .

[7]  C. McDiarmid Concentration , 1862, The Dental register.

[8]  I. Pinelis On normal domination of (super)martingales , 2005, math/0512382.

[9]  Houman Owhadi,et al.  Optimal Uncertainty Quantification , 2010, SIAM Rev..

[10]  W. Hoeffding Probability Inequalities for sums of Bounded Random Variables , 1963 .

[11]  F. Topsøe BOUNDS FOR ENTROPY AND DIVERGENCE FOR DISTRIBUTIONS OVER A TWO-ELEMENT SET , 2001 .

[12]  Lawrence K. Saul,et al.  Large Deviation Methods for Approximate Probabilistic Inference , 1998, UAI.

[13]  Emmanuel Rio,et al.  Local invariance principles and their application to density estimation , 1994 .

[14]  E. Rio,et al.  Inégalités de concentration pour les processus empiriques de classes de parties , 2001 .

[15]  M. Sion On general minimax theorems , 1958 .

[16]  Gustavo L. Gilardoni An Improvement on Vajda’s Inequality , 2008 .

[17]  O. Krafft A note on exponential bounds for binomial probabilities , 1969 .

[18]  V. Bentkus On Hoeffding’s inequalities , 2004, math/0410159.