PGTuner: An Efficient Framework for Automatic and Transferable Configuration Tuning of Proximity Graphs
Hao Duan, Yitong Song, Bin Yao, Anqi Liang

TL;DR
PGTuner is a novel framework that leverages pre-training and transfer learning to efficiently optimize proximity graph configurations for approximate nearest neighbor search, adapting to dynamic datasets and accuracy requirements.
Contribution
It introduces a pre-trained query performance prediction model and a reinforcement learning-based configuration recommendation system for automatic, transferable tuning of proximity graphs.
Findings
Achieves up to 14.69X tuning efficiency improvement.
Maintains top-level tuning performance across diverse datasets.
Enhances dynamic scenario tuning with a 14.64X speedup.
Abstract
Approximate Nearest Neighbor Search (ANNS) plays a crucial role in many key areas. Proximity graphs (PGs) are the leading method for ANNS, offering the best balance between query efficiency and accuracy. However, their performance heavily depends on various construction and query parameters, which are difficult to optimize due to their complex inter-dependencies. Given that users often prioritize specific accuracy levels, efficiently identifying the optimal PG configurations to meet these targets is essential. Although some studies have explored automatic configuration tuning for PGs, they are limited by inefficiencies and suboptimal results. These issues stem from the need to construct numerous PGs for searching and re-tuning from scratch whenever the dataset changes, as well as the failure to capture the complex dependencies between configurations, query performance, and tuning…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
