Deep Spatial Domain Generalization

TL;DR

This paper introduces a novel spatial domain generalization framework using spatial interpolation graph neural networks to handle spatial heterogeneity and predict for unseen locations, validated on multiple real-world datasets.

Contribution

It proposes a generic framework with a spatial interpolation graph neural network to address spatial heterogeneity and unseen location prediction in spatial domain generalization.

Findings

Effective on thirteen real-world datasets

Outperforms existing methods in spatial domain generalization

Successfully predicts for unseen spatial locations

Abstract

Spatial autocorrelation and spatial heterogeneity widely exist in spatial data, which make the traditional machine learning model perform badly. Spatial domain generalization is a spatial extension of domain generalization, which can generalize to unseen spatial domains in continuous 2D space. Specifically, it learns a model under varying data distributions that generalizes to unseen domains. Although tremendous success has been achieved in domain generalization, there exist very few works on spatial domain generalization. The advancement of this area is challenged by: 1) Difficulty in characterizing spatial heterogeneity, and 2) Difficulty in obtaining predictive models for unseen locations without training data. To address these challenges, this paper proposes a generic framework for spatial domain generalization. Specifically, We develop the spatial interpolation graph neural network that handles spatial data as a graph and learns the spatial embedding on each node and their relationships. The spatial interpolation graph neural network infers the spatial embedding of an unseen location during the test phase. Then the spatial embedding of the target location is used to decode the parameters of the downstream-task model directly on the target location. Finally, extensive experiments on thirteen real-world datasets demonstrate the proposed method's strength.