Graph-Based Nearest Neighbor Search: From Practice to Theory
Abstract
Graph-based approaches are empirically shown to be very successful for the nearest neighbor search (NNS). However, there has been very little research on their theoretical guarantees. We fill this gap and rigorously analyze the performance of graph-based NNS algorithms, specifically focusing on the low-dimensional ($d \ll \log n$) regime. In addition to the basic greedy algorithm on nearest neighbor graphs, we also analyze the most successful heuristics commonly used in practice: speeding up via adding shortcut edges and improving accuracy via maintaining a dynamic list of candidates. We believe that our theoretical insights supported by experimental analysis are an important step towards understanding the limits and benefits of graph-based NNS algorithms.
Cite
Text
Prokhorenkova and Shekhovtsov. "Graph-Based Nearest Neighbor Search: From Practice to Theory." International Conference on Machine Learning, 2020.Markdown
[Prokhorenkova and Shekhovtsov. "Graph-Based Nearest Neighbor Search: From Practice to Theory." International Conference on Machine Learning, 2020.](https://mlanthology.org/icml/2020/prokhorenkova2020icml-graphbased/)BibTeX
@inproceedings{prokhorenkova2020icml-graphbased,
title = {{Graph-Based Nearest Neighbor Search: From Practice to Theory}},
author = {Prokhorenkova, Liudmila and Shekhovtsov, Aleksandr},
booktitle = {International Conference on Machine Learning},
year = {2020},
pages = {7803-7813},
volume = {119},
url = {https://mlanthology.org/icml/2020/prokhorenkova2020icml-graphbased/}
}