A Comprehensive Survey and Experimental Comparison of Graph-Based Approximate Nearest Neighbor Search

TL;DR

This paper provides a comprehensive survey and experimental comparison of 13 graph-based approximate nearest neighbor search algorithms, offering insights into their performance, strengths, and areas for improvement across diverse datasets.

Contribution

It introduces a new taxonomy and pipeline for evaluating graph-based ANNS algorithms, and presents an optimized method that surpasses existing state-of-the-art solutions.

Findings

Identified key strengths and weaknesses of 13 algorithms.

Provided rule-of-thumb recommendations for practitioners.

Developed an optimized algorithm outperforming current methods.

Abstract

Approximate nearest neighbor search (ANNS) constitutes an important operation in a multitude of applications, including recommendation systems, information retrieval, and pattern recognition. In the past decade, graph-based ANNS algorithms have been the leading paradigm in this domain, with dozens of graph-based ANNS algorithms proposed. Such algorithms aim to provide effective, efficient solutions for retrieving the nearest neighbors for a given query. Nevertheless, these efforts focus on developing and optimizing algorithms with different approaches, so there is a real need for a comprehensive survey about the approaches' relative performance, strengths, and pitfalls. Thus here we provide a thorough comparative analysis and experimental evaluation of 13 representative graph-based ANNS algorithms via a new taxonomy and fine-grained pipeline. We compared each algorithm in a uniform test environment on eight real-world datasets and 12 synthetic datasets with varying sizes and characteristics. Our study yields novel discoveries, offerings several useful principles to improve algorithms, thus designing an optimized method that outperforms the state-of-the-art algorithms. This effort also helped us pinpoint algorithms' working portions, along with rule-of-thumb recommendations about promising research directions and suitable algorithms for practitioners in different fields.