arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 1401.0212math.OC
收藏

数据驱动的稳健Optimization

Data-Driven Robust Optimization

Dimitris Bertsimas, Vishal Gupta, Nathan Kallus

摘要:

在过去的十年里,运筹学应用程序的数据可用性出现了爆炸式增长。出于这种日益增长的可用性,我们提出了一种新的模式,利用数据来设计不确定性集强大的优化使用统计假设测试。该方法是灵活的,广泛适用的,和强大的最优化问题建立从我们的新的集合是计算上易于处理的,无论是理论上还是实践中。此外,这些问题的最优解具有强的有限样本概率保证。\edit {我们描述了为给定应用选择适当集合的具体过程,并将我们的方法应用于多个不确定约束。投资组合管理和排队的计算证据证实,我们的数据驱动集显着优于传统的强大的优化技术,只要数据可用。

英文摘要:

The last decade witnessed an explosion in the availability of data for operations research applications. Motivated by this growing availability, we propose a novel schema for utilizing data to design uncertainty sets for robust optimization using statistical hypothesis tests. The approach is flexible and widely applicable, and robust optimization problems built from our new sets are computationally tractable, both theoretically and practically. Furthermore, optimal solutions to these problems enjoy a strong, finite-sample probabilistic guarantee. \edit{We describe concrete procedures for choosing an appropriate set for a given application and applying our approach to multiple uncertain constraints. Computational evidence in portfolio management and queuing confirm that our data-driven sets significantly outperform traditional robust optimization techniques whenever data is available.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑