Uniform Inductive Spatio-Temporal Kriging

TL;DR

UniSTOK is a flexible framework that improves spatio-temporal signal prediction from incomplete data by emphasizing reliable observations and calibrating systematic biases, outperforming existing methods.

Contribution

It introduces Reliability-guided Signal Regulation and Residual Bias Calibration to enhance inductive spatio-temporal kriging under incomplete observations.

Findings

UniSTOK consistently outperforms multiple kriging backbones on real-world datasets.

The framework effectively emphasizes reliable data and calibrates systematic biases.

Experimental results demonstrate significant improvements in prediction accuracy.

Abstract

Inductive spatio-temporal kriging infers signals at unobserved locations from observed sensors, but real-world observations are often incomplete and exhibit block-wise missingness caused by failures, interruptions, or maintenance. A common impute-then-krige pipeline suffers from objective mismatch: better reconstruction on observed sensors does not necessarily improve downstream kriging, and value-dependent imputation bias can be propagated to unobserved nodes. We propose UniSTOK, a plug-and-play framework for inductive spatio-temporal kriging under incomplete observations. We first introduce Reliability-guided Signal Regulation (RSR), which estimates entry-wise reliability from temporal continuity and spatial support, and uses it to regulate the input signals so that reliable observations are emphasized while long-gap or weakly supported entries are suppressed before spatial propagation. We further introduce Residual Bias Calibration (RBC), which estimates value-conditioned residual prototypes after the main predictor converges and learns context-correction amplitudes to adaptively calibrate systematic over- or under-estimation in final kriging predictions. Extensive experiments on real-world datasets show that UniSTOK consistently improves multiple kriging backbones.