Adaptive importance sampling for extreme quantile estimation with stochastic black box computer models

Adaptive importance sampling for extreme quantile estimation with stochastic black box computer models
复制标题

使用随机黑盒计算机模型进行极端分位数估计的自适应重要性采样

DOI:
10.1002/nav.21938
复制
发表时间:
2020
期刊:
Naval Research Logistics (NRL
影响因子:
--
通讯作者:
Lam, Henry
Lam, Henry
中科院分区:
--
文献类型:
--
作者:
Pan, Qiyun;Byon, Eunshin;Ko, Young Myoung;Lam, Henry

文献摘要

参考文献

被引文献

相似文献

分位数是可靠性分析中的一个重要量,它与定义失效事件的抗力水平有关。本研究发展一种计算效率高的抽样方法,利用随机黑箱计算机模型估计极端分位数。重要抽样作为一种有效的方差缩减技术,在可靠性研究中被广泛应用,以减少估计的不确定性,提高计算效率。然而,当应用于分位数估计时,重要性抽样面临挑战,因为重要性抽样密度的良好选择依赖于未知分位数的信息。我们提出了一种自适应的方法,细化的重要性采样密度参数对未知的目标分位数值沿着迭代。所提出的自适应方案允许我们使用在以前的迭代中获得的模拟结果,用于引导模拟过程集中在重要的输入区域。我们证明了所提出的方法的一些收敛性,并表明我们的方法可以实现降低方差粗蒙特卡罗采样。通过数值算例和涡轮机的算例分析,验证了该方法的有效性。
Quantile is an important quantity in reliability analysis, as it is related to the resistance level for defining failure events. This study develops a computationally efficient sampling method for estimating extreme quantiles using stochastic black box computer models. Importance sampling has been widely employed as a powerful variance reduction technique to reduce estimation uncertainty and improve computational efficiency in many reliability studies. However, when applied to quantile estimation, importance sampling faces challenges, because a good choice of the importance sampling density relies on information about the unknown quantile. We propose an adaptive method that refines the importance sampling density parameter toward the unknown target quantile value along the iterations. The proposed adaptive scheme allows us to use the simulation outcomes obtained in previous iterations for steering the simulation process to focus on important input areas. We prove some convergence properties of the proposed method and show that our approach can achieve variance reduction over crude Monte Carlo sampling. We demonstrate its estimation efficiency through numerical examples and wind turbine case study.
DOI: 10.1109/wsc.2012.6465096
发表时间: 2012-12
期刊: Proceedings Title: Proceedings of the 2012 Winter Simulation Conference (WSC)
影响因子: --
作者:
X. Chen;B. Nelson;K. Kim
通讯作者: X. Chen;B. Nelson;K. Kim
DOI: 10.1016/j.spl.2014.06.007
发表时间: 2014-10
影响因子: 0.8
作者:
M. Kohler;A. Krzyżak;Harro Walk
通讯作者: M. Kohler;A. Krzyżak;Harro Walk
具有自适应重要性采样的分位数估计
DOI: --
发表时间: 2010
期刊:
影响因子: --
作者:
Daniel Egloff;Markus Leippold
通讯作者: Markus Leippold
DOI: 10.1287/mnsc.1110.1330
发表时间: 2011-06-01
期刊: MANAGEMENT SCIENCE
影响因子: 5.4
作者:
Broadie, Mark;Du, Yiping;Moallemi, Ciamac C.
通讯作者: Moallemi, Ciamac C.
DOI: 10.1007/s11009-017-9599-7
发表时间: 2017
影响因子: 0.9
作者:
Choe, Youngjun;Lam, Henry;Byon, Eunshin
通讯作者: Byon, Eunshin