Efficiently Answering Durability Prediction Queries

Efficiently Answering Durability Prediction Queries
复制标题

高效回答耐久性预测查询

DOI:
10.1145/3448016.3457305
复制
发表时间:
2021
期刊:
Proceedings of the 2021 ACM SIGMOD International Conference on Management of Data
影响因子:
--
通讯作者:
Yang, Jun
Yang, Jun
中科院分区:
--
文献类型:
--
作者:
Gao, Junyang;Xu, Yifan;Agarwal, Pankaj K.;Yang, Jun

文献摘要

参考文献

相似文献

我们考虑一类称为耐久性预测查询的查询,这些查询通常出现在预测分析中,我们使用给定的预测模型来回答有关可能的未来的问题,以告知我们的决策。耐久性预测查询的示例包括“该金融产品在未来12个季度中保持亏损的概率是多少?以及“我们建议的服务器群集在服务级别协议期限结束前未能达到所需服务级别协议的可能性有多大?“我们设计了一种称为多级分割采样(MLSS)的通用方法,可以有效地处理复杂的查询和复杂的模型-包括那些涉及黑盒函数的模型-只要模型允许我们一步一步地模拟可能的未来。我们的方法解决了标准蒙特卡罗(MC)方法的效率低下,通过应用重要性分裂的想法,让一个“有前途的”样本路径前缀生成多个“后代”路径,从而指导模拟工作更有前途的路径。我们提出了实用的技术设计拆分策略,使用户从手动调整。实验表明,我们的方法是能够实现无偏估计和相同的误差保证标准MC,同时提供了一个数量级的成本降低。
We consider a class of queries called durability prediction queries that arise commonly in predictive analytics, where we use a given predictive model to answer questions about possible futures to inform our decisions. Examples of durability prediction queries include "what is the probability that this financial product will keep losing money over the next 12 quarters before turning in any profit?" and "what is the chance for our proposed server cluster to fail the required service-level agreement before its term ends?" We devise a general method called Multi-Level Splitting Sampling (MLSS) that can efficiently handle complex queries and complex models---including those involving black-box functions---as long as the models allow us to simulate possible futures step by step. Our method addresses the inefficiency of standard Monte Carlo (MC) methods by applying the idea of importance splitting to let one "promising" sample path prefix generate multiple "offspring" paths, thereby directing simulation efforts toward more promising paths. We propose practical techniques for designing splitting strategies, freeing users from manual tuning. Experiments show that our approach is able to achieve unbiased estimates and the same error guarantees as standard MC while offering an order-of-magnitude cost reduction.
随着时间的推移一致的 Top-k 查询
DOI: 10.1007/978-3-642-00887-0_5
发表时间: 2009
期刊: --
影响因子: --
作者:
M. Lee;W. Hsu;Ling Li;W. Tok
通讯作者: W. Tok
DOI: 10.14778/3342263.3342642
发表时间: 2019-07-01
影响因子: 2.5
作者:
Gao, Junyang;Li, Xian;Yang, Jun
通讯作者: Yang, Jun
DOI: 10.1145/2020408.2020601
发表时间: 2011-08
期刊: --
影响因子: --
作者:
Xiao Jiang;Chengkai Li;Ping Luo;Min Wang;Yong Yu
通讯作者: Xiao Jiang;Chengkai Li;Ping Luo;Min Wang;Yong Yu
RESTART:快速模拟罕见事件的简单方法
DOI: --
发表时间: 1994
期刊: Proceedings of Winter Simulation Conference
影响因子: --
作者:
M. Villén;J. Villén
通讯作者: J. Villén
DOI: --
发表时间: 2017
影响因子: 2.5
作者:
Neil G. Marchant;Benjamin I. P. Rubinstein
通讯作者: Benjamin I. P. Rubinstein