DeepAI AI Chat
Log In Sign Up

The RLR-Tree: A Reinforcement Learning Based R-Tree for Spatial Data

by   Tu Gu, et al.

Learned indices have been proposed to replace classic index structures like B-Tree with machine learning (ML) models. They require to replace both the indices and query processing algorithms currently deployed by the databases, and such a radical departure is likely to encounter challenges and obstacles. In contrast, we propose a fundamentally different way of using ML techniques to improve on the query performance of the classic R-Tree without the need of changing its structure or query processing algorithms. Specifically, we develop reinforcement learning (RL) based models to decide how to choose a subtree for insertion and how to split a node, instead of relying on hand-crafted heuristic rules as R-Tree and its variants. Experiments on real and synthetic datasets with up to 100 million spatial objects clearly show that our RL based index outperforms R-Tree and its variants.


page 1

page 2

page 3

page 4


Efficient Spatial Nearest Neighbor Queries Based on Multi-layer Voronoi Diagrams

Nearest neighbor (NN) problem is an important scientific problem. The NN...

The "AI+R"-tree: An Instance-optimized R-tree

The emerging class of instance-optimized systems has shown potential to ...

GLIN: A Lightweight Learned Indexing Mechanism for Complex Geometries

Although spatial index structures shorten the query response time, they ...

Learning Augmented Binary Search Trees

A treap is a classic randomized binary search tree data structure that i...

Index-based Solutions for Efficient Density Peaks Clustering

Density Peaks Clustering (DPC), a novel density-based clustering approac...

QDR-Tree: An Efcient Index Scheme for Complex Spatial Keyword Query

With the popularity of mobile devices and the development of geo-positio...

Aggregate Analytic Window Query over Spatial Data

Analytic window query is a commonly used query in the relational databas...