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 1910.03534cs.IRcs.LG
收藏

基于邻域图的复杂非度量数据快速准确检索

Accurate and Fast Retrieval for Complex Non-metric Data via Neighborhood Graphs

Carnegie Mellon University(卡内基梅隆大学)

Leonid Boytsov, Eric Nyberg

摘要:

我们证明了一个基于图的搜索算法,依赖于一个近似的邻域图的建设,可以直接与具有挑战性的非度量和/或非对称的距离,而不诉诸度量空间映射和/或距离对称化,这反过来又会导致大量的性能下降。虽然直接的度量化和对称化通常是无效的,但我们发现使用修改的,例如,对称的距离可以提高性能。这一观察为设计特定于索引的图构造距离函数的新研究路线铺平了道路。

英文摘要:

We demonstrate that a graph-based search algorithm-relying on the construction of an approximate neighborhood graph-can directly work with challenging non-metric and/or non-symmetric distances without resorting to metric-space mapping and/or distance symmetrization, which, in turn, lead to substantial performance degradation. Although the straightforward metrization and symmetrization is usually ineffective, we find that constructing an index using a modified, e.g., symmetrized, distance can improve performance. This observation paves a way to a new line of research of designing index-specific graph-construction distance functions.

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。

↑