UniSTOK: Uniform Inductive Spatio-Temporal Kriging

2026-03-05

AI Summary

This work addresses three key challenges in spatiotemporal kriging—difficulty in distinguishing true from spurious signals due to high heterogeneity, complex missing patterns, and distortion of local structures—by proposing UniSTOK, a plug-and-play framework. UniSTOK employs a dual-branch architecture that processes both original observations and puzzle-augmented surrogate signals in parallel, leveraging a shared spatiotemporal backbone network with explicit missing-mask modulation. A dual-channel attention mechanism adaptively fuses features from both branches, enabling robust representation learning. The framework is compatible with existing inductive models and demonstrates significant improvements in inference accuracy and robustness across diverse real-world datasets and under complex missing-data scenarios.

Technology Category

• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning
• Data Mining & Knowledge Management: Mining of Spatial, Temporal or Spatio-Temporal Data
• Reasoning under Uncertainty: Stochastic Optimization

Application Category

• Graph Algorithms and Modeling for the Web: Algorithms and analysis for incomplete, noisy, or partially observed Web-related graphs
• Semantics and Knowledge: Methods, algorithms and applications for the development of semantic models, knowledge graphs and other forms of structured data models with machine-interpretable semantics
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Spatio-temporal kriging aims to infer signals at unobserved locations from observed sensors and is critical to applications such as transportation and environmental monitoring. In practice, however, observed sensors themselves often exhibit heterogeneous missingness, forcing inductive kriging models to rely on crudely imputed inputs. This setting brings three key challenges: (1) it is unclear whether an value is a true signal or a missingness-induced artifact; (2) missingness is highly heterogeneous across sensors and time; (3) missing observations distort the local spatio-temporal structure. To address these issues, we propose Uniform Inductive Spatio-Temporal Kriging (UniSTOK), a plug-and-play framework that enhances existing inductive kriging backbones under missing observation. Our framework forms a dual-branch input consisting of the original observations and a jigsaw-augmented counterpart that synthesizes proxy signals only at missing entries. The two branches are then processed in parallel by a shared spatio-temporal backbone with explicit missingness mask modulation. Their outputs are finally adaptively fused via dual-channel attention. Experiments on multiple real-world datasets under diverse missing patterns demonstrate consistent and significant improvements.

Problem

• spatio-temporal kriging
• missing data
• heterogeneous missingness
• inductive modeling
• sensor networks

Innovation

• inductive kriging
• missing data
• spatio-temporal modeling
• jigsaw augmentation
• dual-branch attention