Almost Navigable Graphs

2026-07-16

AI Summary

This work addresses the inefficiency of traditional navigable graphs in large-scale vector data, where excessive edge counts hinder both construction and search. The authors introduce the novel concept of “γ-approximate navigability,” formally defining it and proving that such graphs require only O(n/(1−γ)) edges, substantially reducing graph density. Building upon this theoretical foundation, they devise a near-linear-time randomized construction algorithm coupled with a greedy beam search strategy, enabling efficient approximate nearest neighbor retrieval with provable guarantees. Empirical evaluations across multiple billion-scale datasets demonstrate that the proposed method achieves superior recall–latency trade-offs while using less than half the average degree of conventional graphs, significantly reducing memory consumption and accelerating search.

Abstract

Graph-based methods like HNSW, DiskANN, NSG, and others have become an increasingly popular choice for implementing approximate nearest neighbor search (ANNS) in Vector Databases (VecDBs). The success of these methods has motivated the study of how to best construct a search graph for a given dataset. To that end, \emph{navigability} has been identified as a desirable graph property which ensures good ANNS performance when combined with greedy search. However, for a dataset with $n$ vectors, the sparsest navigable graph requires $O(n\sqrt{n})$ edges in the worst-case, and we show empirically that, for typical billion node datasets, 100s of edges are needed per node. This leads to slow search and high memory requirements. Moreover, under standard complexity theoretical assumptions, it was recently established that constructing a sparse navigable graph requires $Ω(n^{2-ε})$ time, which is prohibitive for large datasets. We address these concerns by introducing a relaxed notation of navigability called ``$γ$-almost navigability'' for any $γ\in [0,1]$, with $γ= 1$ corresponding to full navigability. We prove that any dataset (under any distance) admits a $γ$-almost navigable graph with just $O\left(\frac{n}{1-γ}\right)$ edges, linear in the dataset size. We present a randomized algorithm for constructing such a graph in near-linear time. While we prove that $γ$-almost navigability sacrifices the worst-case search guarantees enjoyed by navigability, we show empirically that greedy beam search still performs well in such graphs when $γ< 1$. Indeed, we obtain improved recall-runtime tradeoffs on a variety of datasets compared to fully navigable graphs. Moreover, our graphs are more space efficient, with degree typically less than half that of a fully navigable graph for comparable performance.

Problem

• approximate nearest neighbor search
• navigability
• graph construction
• vector databases
• scalability

Innovation

• almost navigability
• approximate nearest neighbor search
• graph sparsification
• vector databases
• greedy beam search