In Defense of Locality-Sensitive Hashing

CASIA OpenIR > 多模态人工智能系统全国重点实验室 > 先进时空数据分析与学习

	In Defense of Locality-Sensitive Hashing
	Kun Ding; Chunlei Huo; Bin Fan; Shiming Xiang; Chunhong Pan
发表期刊	IEEE Transactions on Neural Networks and Learning Systems
	2018-01
卷号	29 期号:1 页码:87-103
摘要	Hashing-based semantic similarity search is becoming increasingly important for building large-scale content-based retrieval system. The state-of-the-art supervised hashing techniques use flexible two-step strategy to learn hash functions. The first step learns binary codes for training data by solving binary optimization problems with millions of variables, thus usually requiring intensive computations. Despite simplicity and efficiency, locality-sensitive hashing (LSH) has never been recognized as a good way to generate such codes due to its poor performance in traditional approximate neighbor search. We claim in this paper that the true merit of LSH lies in transforming the semantic labels to obtain the binary codes, resulting in an effective and efficient two-step hashing framework. Specifically, we developed the locality-sensitive two-step hashing (LS-TSH) that generates the binary codes through LSH rather than any complex optimization technique. Theoretically, with proper assumption, LS-TSH is actually a useful LSH scheme, so that it preserves the label-based semantic similarity and possesses sublinear query complexity for hash lookup. Experimentally, LS-TSH could obtain comparable retrieval accuracy with state of the arts with two to three orders of magnitudes faster training speed.
关键词	Locality-sensitive Hashing (Lsh) Semantic Similarity Search Two-step Hashing
文献类型	期刊论文
条目标识符	http://ir.ia.ac.cn/handle/173211/20366
专题	多模态人工智能系统全国重点实验室_先进时空数据分析与学习
作者单位	National Laboratory of Pattern Recognition, Institute of Automation, Chinese Academy of Sciences
第一作者单位	模式识别国家重点实验室
推荐引用方式 GB/T 7714	Kun Ding,Chunlei Huo,Bin Fan,et al. In Defense of Locality-Sensitive Hashing[J]. IEEE Transactions on Neural Networks and Learning Systems,2018,29(1):87-103.
APA	Kun Ding,Chunlei Huo,Bin Fan,Shiming Xiang,&Chunhong Pan.(2018).In Defense of Locality-Sensitive Hashing.IEEE Transactions on Neural Networks and Learning Systems,29(1),87-103.
MLA	Kun Ding,et al."In Defense of Locality-Sensitive Hashing".IEEE Transactions on Neural Networks and Learning Systems 29.1(2018):87-103.

条目包含的文件		下载所有文件
文件名称/大小	文献类型	版本类型	开放类型	使用许可
07636996.pdf（2624KB）	期刊论文	作者接受稿	开放获取	CC BY-NC-SA	浏览下载