高效的基于流的最大最小多样化方法及其最小化失败率
Efficient stream-based Max-Min diversification with minimal failure rate
AI总结：
本文针对流式最大最小多样化问题，提出基于秩的失败率最小化算法，显著降低选择失败概率，并通过模拟验证其性能优于现有策略。
摘要：
流式最大最小多样化问题关注从已知有限长度的数据流中选择一个有限且多样化的样本项。最大化的目标是在任意一对被选中的项之间的最小距离。我们考虑不可撤销选择采样，即在处理流中的项时需要立即且不可撤销地做出决定，这是一个在文献中很少被研究的设置。标准的顺序选择算法方法忽略了选择失败，即当流的最后一些项被默认选取时，以避免交付不完整的选择集。这个缺陷对于最大最小多样化目标可能是灾难性的。所提出的失败率最小化（FRM）是一种基于秩的算法，它选择一组多样化的项，并且显著降低了失败的概率。我们通过模拟展示了FRM与现有选择策略相比的性能。
英文摘要：
The streaming max-min diversification problem concerns the selection of a limited and diverse sample of items out of a data stream of known finite length. The objective to be maximized is the minimum distance among any pair of selected items. We consider the irrevocable-choice sampling, where decisions need to be immediate and irrevocable while processing the items of the stream, which is a setting little studied in the literature. Standard algorithmic approaches for sequential selection disregard selection failures, which is when the last items of the stream are picked by default, to prevent delivering an incomplete selection set. This defect can be catastrophic for the max-min diversification objective. The proposed Failure Rate Minimization (FRM) is a rank-based algorithm that selects a set of diverse items and, in addition, reduces significantly the probability of having failures. We demonstrate with simulations FRM's performance comparing with existing selection strategies.
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。
