← Back to all news
arXiv Machine Learning By Minghao Li, Raghav Mittal, Sanjivni Rana, Suraj Shetiya, Gautam Das, Nick Koudas

HNSW with Accuracy Guarantees Using Graph Spanners -- A Technical Report

Read the original on arXiv Machine Learning →

arXiv:2607. 02338v1 Announce Type: cross Abstract: Hierarchical Navigable Small World (HNSW) graphs serve as the industry standard due to their logarithmic complexity and strong empirical performance.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Machine Learning.

Related stories

arXiv Machine Learning
Jul 7

HNSW with Accuracy Guarantees Using Graph Spanners

arXiv:2607. 02338v2 Announce Type: replace-cross Abstract: Hierarchical Navigable Small World (HNSW) graphs serve as the industry standard due to their logarithmic complexity and strong empirical performance.

By Minghao Li, Raghav Mittal, Sanjivni Rana, Suraj Shetiya, Gautam Das, Nick Koudas
benchmarks
More like this →
arXiv AI
Jun 16

Adaptive $k$NN graph model

arXiv:2601. 16509v2 Announce Type: replace-cross Abstract: The $k$-nearest neighbors ($k$NN) algorithm is a cornerstone of non-parametric classification in artificial intelligence, yet its deployment in large-scale applications is persistently constrained by the computational trade-off between inference speed and accuracy.

By Jiaye Li, Hang Xu, Shichao Zhang
efficiencybenchmarks
More like this →
arXiv AI
Aug 19

Average Distance Approximation for Static Large Graphs

The paper investigates methods for estimating average distances in large static graphs, comparing a graph sampling approach (Random Walk) with landmark-based techniques such as the Size Estimation Framework (SEF) and the Eppstein‑Wang (EW) algorithm. Random Walk proved unreliable for small samples and costly for larger ones, while landmark methods using HyperLogLog were more efficient. Experiments on undirected, unweighted graphs showed that the EW algorithm achieves very low error (≈0.02%) and can use a subset of only 100 nodes for accurate estimation, outperforming SEF in accuracy and speed.

By Kartikey Ahlawat
More like this →
arXiv Machine Learning
Jun 18

Compact Geometric Representations of Hierarchies

arXiv:2606. 18520v1 Announce Type: cross Abstract: Computing geometric representations of data is a cornerstone of modern machine learning, typically achieved by training dual encoders which map queries and documents into a shared embedding space.

By Prashant Gokhale, Piotr Indyk, Yuhao Liu, Sandeep Silwal, Tony Chang Wang, Haike Xu
rag
More like this →