SpaceTime: Causal Discovery from Non-Stationary Time Series

2025-01-17

AI Summary

Existing causal discovery methods struggle to jointly model temporal change points, spatial heterogeneity, and invariant causal mechanisms in nonstationary, spatially heterogeneous multivariate time series. Method: We propose a unified framework that jointly solves three tasks: causal graph discovery, temporal segmentation (change-point detection), and spatiotemporal homogeneous region partitioning. Our approach is the first to integrate spatial heterogeneity and temporal nonstationarity within a single causal modeling framework. We design a Minimum Description Length (MDL)-based consistent scoring criterion supporting nonparametric functional modeling and kernel-based heterogeneity testing. Contribution/Results: The method automatically identifies spatiotemporal subregions where the causal mechanism remains invariant. Evaluated on multi-basin streamflow and cross-ecosystem flux datasets, it successfully recovers physically plausible time-varying causal structures and ecologically meaningful partitions, significantly outperforming stationary baselines.

Technology Category

• Machine Learning: Causal Learning
• Reasoning under Uncertainty: Causality
• Data Mining & Knowledge Management: Mining of Spatial, Temporal or Spatio-Temporal Data

Application Category

• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• Web Mining and Content Analysis: Models for Web evolution
• 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

Abstract

Understanding causality is challenging and often complicated by changing causal relationships over time and across environments. Climate patterns, for example, shift over time with recurring seasonal trends, while also depending on geographical characteristics such as ecosystem variability. Existing methods for discovering causal graphs from time series either assume stationarity, do not permit both temporal and spatial distribution changes, or are unaware of locations with the same causal relationships. In this work, we therefore unify the three tasks of causal graph discovery in the non-stationary multi-context setting, of reconstructing temporal regimes, and of partitioning datasets and time intervals into those where invariant causal relationships hold. To construct a consistent score that forms the basis of our method, we employ the Minimum Description Length principle. Our resulting algorithm SPACETIME simultaneously accounts for heterogeneity across space and non-stationarity over time. Given multiple time series, it discovers regime changepoints and a temporal causal graph using non-parametric functional modeling and kernelized discrepancy testing. We also show that our method provides insights into real-world phenomena such as river-runoff measured at different catchments and biosphere-atmosphere interactions across ecosystems.

Problem

• Causal inference
• Time-varying effects
• Location-specific impacts

Innovation

• SPACETIME algorithm
• causal relationship dynamics
• Minimum Description Length principle