Efficient simulation for branching linear recursions

Efficient simulation for branching linear recursions
复制标题

分支线性递归的高效模拟

DOI:
10.1109/wsc.2015.7408378
复制
发表时间:
2015
期刊:
2015 Winter Simulation Conference (WSC)
影响因子:
--
通讯作者:
Mariana Olvera
Mariana Olvera
中科院分区:
--
文献类型:
--
作者:
Ningyuan Chen;Mariana Olvera

文献摘要

参考文献

被引文献

相似文献

给出了一种模拟线性分支分布方程唯一吸引不动点的算法。这样的方程出现在信息排序算法的分析中,如PageRank,也出现在分治算法的复杂性分析中,如Quicksort。朴素的模拟方法是精确地模拟加权分支过程的适当代数,该过程在采样的代数中具有指数级的复杂性。相反,我们提出了一种具有线性复杂度的迭代自举算法;在此基础上证明了它的收敛性和一组估计量的相合性。
We provide an algorithm for simulating the unique attracting fixed-point of linear branching distributional equations. Such equations appear in the analysis of information ranking algorithms, e.g., PageRank, and in the complexity analysis of divide and conquer algorithms, e.g., Quicksort. The naive simulation approach would be to simulate exactly a suitable number of generations of a weighted branching process, which has exponential complexity in the number of generations being sampled. Instead, we propose an iterative bootstrap algorithm that has linear complexity; we prove its convergence and the consistency of a family of estimators based on our approach.
DOI: 10.1214/11-aop670
发表时间: 2010
影响因子: 2.3
作者:
Alsmeyer;Gerold ;Biggins;Meiners;Matthias
通讯作者: Matthias
DOI: 10.1080/10236198.2011.589514
发表时间: 2010
影响因子: 1.1
作者:
Alsmeyer;Gerold ;Meiners;Matthias
通讯作者: Matthias