学习型过滤器综述
Survey of Learning-based Filters
李猛 1戴海鹏 1眭永熙 1顾荣 1陈贵海1
作者信息
- 1. 计算机软件新技术国家重点实验室(南京大学) 南京 210023
- 折叠
摘要
作为一种高效的概率性结构,过滤器可以高效地解决近似集合成员查询问题.近年来,随着机器学习技术的发展,一些学习型过滤器表现出色,超越了传统的过滤器.这些学习型过滤器考虑数据分布信息,将集合成员查询问题视为二分类问题,实现了超越传统过滤器的性能.受此启发,学习型过滤器研究领域迅速发展,出现了多个变种.然而,目前还缺乏对近些年相关工作的系统性回顾和比较.为了填补上述空缺,文中全面回顾了近年来的学习型过滤器相关工作,并展望了未来的发展方向.
Abstract
As a space-efficient probability structure,filters can efficiently solve approximate set membership queries.In recent years,with the development of machine learning technology,some learning-based filters have exceeded traditional filters in per-formance.These learning-based filters propose to consider data distribution information and treat set membership queries as a bi-nary classification problem,achieving superior performance compared to traditional filters.Inspired by this,the research field of learning-based filters has progressed rapidly,and several variants have emerged.However,there is still a lack of a systematic re-view and comparison of recent related work.In order to fill this gap,this paper comprehensively reviews recent related work on learning-based filters,analyzes their structure design and theoretical analysis,and predicts the future development direction.
关键词
近似成员资格查询/机器学习/Bloom/过滤器/学习型过滤器/假阳率Key words
Approximate membership query/Machine learning/Bloom filter/Learning-based filter/False positive rate引用本文复制引用
出版年
2024
NETL
NSTL
万方数据