ANN-Benchmarks: A Benchmarking Tool for Approximate Nearest Neighbor Algorithms

被引:80
作者
Aumueller, Martin [1 ]
Bernhardsson, Erik [2 ]
Faithfull, Alexander [1 ]
机构
[1] IT Univ Copenhagen, Copenhagen, Denmark
[2] Better, New York, NY USA
来源
SIMILARITY SEARCH AND APPLICATIONS, SISAP 2017 | 2017年 / 10609卷
基金
欧洲研究理事会;
关键词
D O I
10.1007/978-3-319-68474-1_3
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
This paper describes ANN-Benchmarks, a tool for evaluating the performance of in-memory approximate nearest neighbor algorithms. It provides a standard interface for measuring the performance and quality achieved by nearest neighbor algorithms on different standard data sets. It supports several different ways of integrating k-NN algorithms, and its configuration system automatically tests a range of parameter settings for each algorithm. Algorithms are compared with respect to many different (approximate) quality measures, and adding more is easy and fast; the included plotting front-ends can visualise these as images, LATEX plots, and websites with interactive plots. ANN-Benchmarks aims to provide a constantly updated overview of the current state of the art of k-NN algorithms. In the short term, this overview allows users to choose the correct k-NN algorithm and parameters for their similarity search task; in the longer term, algorithm designers will be able to use this overview to test and refine automatic parameter tuning. The paper gives an overview of the system, evaluates the results of the benchmark, and points out directions for future work. Interestingly, very different approaches to k-NN search yield comparable quality-performance trade-offs.
引用
收藏
页码:34 / 49
页数:16
相关论文
共 26 条
[11]  
Dong W., 2008, P ACM C INF KNOWL MA, P669
[12]  
Edel M., 2014, PROC NIPS WORKSHOP S
[13]   Spherical Hashing: Binary Code Embedding with Hyperspheres [J].
Heo, Jae-Pil ;
Lee, Youngwoon ;
He, Junfeng ;
Chang, Shih-Fu ;
Yoon, Sung-Eui .
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 2015, 37 (11) :2304-2316
[14]  
Indyk P., 1998, Proceedings of the Thirtieth Annual ACM Symposium on Theory of Computing, P604, DOI 10.1145/276698.276876
[15]   EXTENSIONS OF LIPSCHITZ-MAPS INTO BANACH-SPACES [J].
JOHNSON, WB ;
LINDENSTRAUSS, J ;
SCHECHTMAN, G .
ISRAEL JOURNAL OF MATHEMATICS, 1986, 54 (02) :129-138
[16]   The (black) art of runtime evaluation: Are we comparing algorithms or implementations? [J].
Kriegel, Hans-Peter ;
Schubert, Erich ;
Zimek, Arthur .
KNOWLEDGE AND INFORMATION SYSTEMS, 2017, 52 (02) :341-378
[17]   Gradient-based learning applied to document recognition [J].
Lecun, Y ;
Bottou, L ;
Bengio, Y ;
Haffner, P .
PROCEEDINGS OF THE IEEE, 1998, 86 (11) :2278-2324
[18]   Efficient and Robust Approximate Nearest Neighbor Search Using Hierarchical Navigable Small World Graphs [J].
Malkov, Yu A. ;
Yashunin, D. A. .
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 2020, 42 (04) :824-836
[19]   Approximate nearest neighbor algorithm based on navigable small world graphs [J].
Malkov, Yury ;
Ponomarenko, Alexander ;
Logvinov, Andrey ;
Krylov, Vladimir .
INFORMATION SYSTEMS, 2014, 45 :61-68
[20]  
Mikolov T., 2013, ADV NEURAL INFORM PR, P3111