English

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

Information Retrieval 2018-07-19 v2 Databases

Abstract

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 kk-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\LaTeX plots, and websites with interactive plots. ANN-Benchmarks aims to provide a constantly updated overview of the current state of the art of kk-NN algorithms. In the short term, this overview allows users to choose the correct kk-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 kk-NN search yield comparable quality-performance trade-offs. The system is available at http://ann-benchmarks.com .

Keywords

Cite

@article{arxiv.1807.05614,
  title  = {ANN-Benchmarks: A Benchmarking Tool for Approximate Nearest Neighbor Algorithms},
  author = {Martin Aumüller and Erik Bernhardsson and Alexander Faithfull},
  journal= {arXiv preprint arXiv:1807.05614},
  year   = {2018}
}

Comments

Full version of the SISAP 2017 conference paper. v2: Updated the abstract to avoid arXiv linking to the wrong URL

R2 v1 2026-06-23T03:02:01.346Z