Fast Approximate Nearest Neighbor Search With The Navigating Spreading-out Graph

TL;DR

This paper introduces the Navigating Spreading-out Graph (NSG), a scalable and efficient graph-based method for approximate nearest neighbor search that significantly outperforms existing algorithms and is suitable for billion-node datasets.

Contribution

The paper proposes the NSG structure, approximating the Monotonic Relative Neighborhood Graph to improve search efficiency and scalability in billion-node ANNS applications.

Findings

NSG achieves near-logarithmic search complexity.

NSG outperforms existing algorithms on large-scale datasets.

NSG is integrated into Alibaba's Taobao search engine.

Abstract

Approximate nearest neighbor search (ANNS) is a fundamental problem in databases and data mining. A scalable ANNS algorithm should be both memory-efficient and fast. Some early graph-based approaches have shown attractive theoretical guarantees on search time complexity, but they all suffer from the problem of high indexing time complexity. Recently, some graph-based methods have been proposed to reduce indexing complexity by approximating the traditional graphs; these methods have achieved revolutionary performance on million-scale datasets. Yet, they still can not scale to billion-node databases. In this paper, to further improve the search-efficiency and scalability of graph-based methods, we start by introducing four aspects: (1) ensuring the connectivity of the graph; (2) lowering the average out-degree of the graph for fast traversal; (3) shortening the search path; and (4) reducing the index size. Then, we propose a novel graph structure called Monotonic Relative Neighborhood Graph (MRNG) which guarantees very low search complexity (close to logarithmic time). To further lower the indexing complexity and make it practical for billion-node ANNS problems, we propose a novel graph structure named Navigating Spreading-out Graph (NSG) by approximating the MRNG. The NSG takes the four aspects into account simultaneously. Extensive experiments show that NSG outperforms all the existing algorithms significantly. In addition, NSG shows superior performance in the E-commercial search scenario of Taobao (Alibaba Group) and has been integrated into their search engine at billion-node scale.