Gaussian averages, Bernoulli averages, and Gibbs' measures

Gaussian averages, Bernoulli averages, and Gibbs' measures
复制标题

高斯平均值、伯努利平均值和吉布斯度量

DOI:
--
复制
发表时间:
2002
期刊:
Random Struct. Algorithms
影响因子:
--
通讯作者:
M. Talagrand
M. Talagrand
中科院分区:
--
文献类型:
--
作者:
M. Talagrand

文献摘要

被引文献

相似文献

本文证明了一个新的比较原理,其中(gi)是独立同分布的,∑i≤ Ngi ai的上确界的期望值之间的比较原理。标准高斯函数的上确界,以及当r. v. gi被Bernoulli r. v. ηi代替时,同一族的上确界的期望值,P(ηi = ±1)= 1/2。其主要思想是通过加权平均来近似上确界,这是自旋玻璃理论中众所周知的想法。© 2002 Wiley Periodicals,Inc.随机结构算法,21:197-204,2002
We prove a new comparison principle between the expected value of the supremum of a family of r.v. of the type ∑i≤N giai, where (gi) are i.i.d. standard gaussian, and the expected value of the supremum of the same family when the r.v. gi are replaced by Bernoulli r.v. ηi, P(ηi = ±1) = 1/2. The main idea is to approximate the supremum by a weighted average, an idea well known in the theory of spin glasses. © 2002 Wiley Periodicals, Inc. Random Struct. Alg., 21: 197–204, 2002