arrow
返回

Scalable neural classifier-based adaptive importance sampling for failure probability estimation

delete2026-09-09
delete0
PRE
AI
T
Tianyu Zhang
J
Jize Zhang *
DOI:10.1016/j.strusafe.2026.102765delete
delete原文链接
delete原文求助
delete分享
delete收藏
摘要

摘要

En 中文
• 神经分类器实现了复杂高维失效边界的建模。 • 朗之万采样指导了重要性采样(IS)提议样本的生成。 • 防御性重要性采样提高了失效概率估计的鲁棒性。 • 在最高达784维基于图像的案例研究中证明了其有效性。
Keyword:
Reliability analysis
Importance sampling
Rare event estimation
Failure probability estimation
High dimensions

期刊

Structural Safety 封面图
Structural Safety
IF:
6.3
论文数:
1.4K
被引数:
7.0K

机构

暂无机构信息
引用论文

引用论文

Multi-point active learning probability density evolution method多点主动学习概率密度演化方法
err2025-07-08
err0
PREAI
errTong Zhou; Tong Guo; Xujia Zhu; Alexandros A. Taflanidis; Jize Zhang
err分享
err收藏
Cross entropy-based importance sampling using Gaussian densities revisited
err2019-01-01
err87
errOAAI
errGeyer, Sebastian; Papaioannou, Iason; Straub, Daniel
err分享
err收藏
err分享
err收藏
err分享
err收藏
err分享
err收藏
Rare event estimation with sequential directional importance sampling
err2023-01-01
err29
PREAI
errCheng, Kai; Papaioannou, Iason; Lu, Zhenzhou; Zhang, Xiaobo; Wang, Yanping
err分享
err收藏
Second-order reliability methods: a review and comparative study
err2021-11-10
err54
errOAAI
errHu, Zhangli; Mansour, Rami; Olsson, Marten; Du, Xiaoping
err分享
err收藏
学者 查看更多内容