Dynamically Detect and Fix Hardness for Efficient Approximate Nearest Neighbor Search

2025-10-25

AI Summary

Graph-based Approximate Nearest Neighbor Search (ANNS) suffers substantial accuracy degradation under out-of-distribution (OOD) queries. Existing approaches like RoarGraph rely on historical queries to construct bipartite graphs but lack theoretical foundations, incur high construction overhead, and poorly adapt to dynamic workloads. Method: We propose a dynamic graph defect detection and repair framework. First, we define *Escape Hardness* to quantify node navigation difficulty and progressively identify and rectify structural defects. Then, we introduce *Reachability Fixing* to enhance critical node accessibility and *Neighboring Graph Defects Fixing* to improve connectivity in high-density query regions. Contribution/Results: Our method requires minimal historical queries, is theoretically grounded, and enables efficient index construction. Experiments on real-world datasets show that, at 99% recall for OOD queries, it achieves 2.25× speedup over baseline search and accelerates index construction by 2.35–9.02×, significantly outperforming RoarGraph and HNSW.

Technology Category

• Search and Optimization: Distributed Search
• Machine Learning: Graph-based Machine Learning
• Data Mining & Knowledge Management: Graph Mining, Social Network Analysis & Community

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Semantics and Knowledge: Scalable techniques for the creation, curation, publication, maintenance, and consumption of large, Web-based, structured, reusable, knowledge graphs and ontologies

Abstract

Approximate Nearest Neighbor Search (ANNS) has become a fundamental component in many real-world applications. Among various ANNS algorithms, graph-based methods are state-of-the-art. However, ANNS often suffers from a significant drop in accuracy for certain queries, especially in Out-of-Distribution (OOD) scenarios. To address this issue, a recent approach named RoarGraph constructs a bipartite graph between the base data and historical queries to bridge the gap between two different distributions. However, it suffers from some limitations: (1) Building a bipartite graph between two distributions lacks theoretical support, resulting in the query distribution not being effectively utilized by the graph index. (2) Requires a sufficient number of historical queries before graph construction and suffers from high construction times. (3) When the query workload changes, it requires reconstruction to maintain high search accuracy. In this paper, we first propose Escape Hardness, a metric to evaluate the quality of the graph structure around the query. Then we divide the graph search into two stages and dynamically identify and fix defective graph regions in each stage based on Escape Hardness. (1) From the entry point to the vicinity of the query. We propose Reachability Fixing (RFix), which enhances the navigability of some key nodes. (2) Searching within the vicinity of the query. We propose Neighboring Graph Defects Fixing (NGFix) to improve graph connectivity in regions where queries are densely distributed. The results of extensive experiments show that our method outperforms other state-of-the-art methods on real-world datasets, achieving up to 2.25x faster search speed for OOD queries at 99% recall compared with RoarGraph and 6.88x faster speed compared with HNSW. It also accelerates index construction by 2.35-9.02x compared to RoarGraph.

Problem

• Detect and fix graph defects in approximate nearest neighbor search
• Improve search accuracy for out-of-distribution query scenarios
• Accelerate graph construction while maintaining high recall performance

Innovation

• Dynamically identifies and fixes defective graph regions
• Enhances navigability using Reachability Fixing (RFix) technique
• Improves connectivity with Neighboring Graph Defects Fixing (NGFix)