A Multivariate Generalization of Hoeffding’s Inequality

نویسنده

  • Péter Major
چکیده

In the study of U -statistics we need a multivariate version of this result. The goal of this paper is to present such an inequality. To formulate it first we have to introduce some notations. Let us fix a positive integer k and some real numbers a(j1, . . . , jk) for all sets of arguments {j1, . . . , jk} such that 1 ≤ jl ≤ n, 1 ≤ l ≤ k, and jl 6= jl′ if l 6= l, in such a way that the numbers a(j1, . . . , jk) are symmetric functions of their arguments, i.e. a(j1, . . . , jk) = a(jπ(1), . . . , jπ(k)) for all permutations π ∈ Πk of the set {1, . . . , k}. Let us define with the help of the above real numbers and a sequence of independent random variables ε1, . . . , εn, P (εj = 1) = P (εj = −1) = 12 , 1 ≤ j ≤ n, the random variable Z = ∑

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

A Multivariate Version of Hoeffding’s In- Equality

In this paper a multivariate version of Hoeffding’s inequality is proved about the tail distribution of homogeneous polynomials of Rademacher functions with an optimal constant in the exponent of the upper bound. The proof is based on an estimate about the moments of homogeneous polynomials of Rademacher functions which can be considered as an improvement of Borell’s inequality in a most import...

متن کامل

CS 269 : Machine Learning Theory Lecture 4 : Infinite Function Classes

Before stating Hoeffding’s Inequality, we recall two intermediate results that we will use in order to prove it. One is Markov’s Inequality and other is Hoeffding’s Lemma. (Note that in class we did not cover Hoeffding’s Lemma, and only gave a brief outline of the Chernoff Bounding Techniques and how they are used to prove Hoeffding’s Inequality. Here we give a full proof of Hoeffding’s Inequal...

متن کامل

Empirical Bernstein Boosting

Concentration inequalities that incorporate variance information (such as Bernstein’s or Bennett’s inequality) are often significantly tighter than counterparts (such as Hoeffding’s inequality) that disregard variance. Nevertheless, many state of the art machine learning algorithms for classification problems like AdaBoost and support vector machines (SVMs) extensively use Hoeffding’s inequalit...

متن کامل

On the generalization of Trapezoid Inequality for functions of two variables with bounded variation and applications

In this paper, a generalization of trapezoid inequality for functions of two independent variables with bounded variation and some applications are given.

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

عنوان ژورنال:

دوره   شماره 

صفحات  -

تاریخ انتشار 2004