🤖 AI Summary
In nonstationary spatiotemporal time series, causal networks are unidentifiable due to Markov equivalence, hindering unique causal structure recovery. Method: We propose the Unique Causal Network (UCN) model, introducing a decomposition-based Higher-Order Causal Entropy (HCE) criterion, integrated with lag-aware modeling and nearest-neighbor entropy estimation for distributed causal structure learning. Contribution/Results: UCN is the first framework to rigorously establish identifiability of causal structures in dynamic spatiotemporal settings. It overcomes fundamental limitations of traditional Bayesian networks and Granger causality—particularly in nonstationary, time-varying, and cyclically dependent regimes—while enabling explicit lag modeling and dynamic causal discovery. Extensive experiments on multiple nonstationary time-series benchmarks demonstrate that UCN consistently outperforms state-of-the-art BN-based and Granger-type methods, achieving new SOTA accuracy in causal structure recovery.
📝 Abstract
Identifying causality is a challenging task in many data-intensive scenarios. Many algorithms have been proposed for this critical task. However, most of them consider the learning algorithms for directed acyclic graph (DAG) of Bayesian network (BN). These BN-based models only have limited causal explainability because of the issue of Markov equivalence class. Moreover, they are dependent on the assumption of stationarity, whereas many sampling time series from complex system are nonstationary. The nonstationary time series bring dataset shift problem, which leads to the unsatisfactory performances of these algorithms. To fill these gaps, a novel causation model named Unique Causal Network (UCN) is proposed in this paper. Different from the previous BN-based models, UCN considers the influence of time delay, and proves the uniqueness of obtained network structure, which addresses the issue of Markov equivalence class. Furthermore, based on the decomposability property of UCN, a higher-order causal entropy (HCE) algorithm is designed to identify the structure of UCN in a distributed way. HCE algorithm measures the strength of causality by using nearest-neighbors entropy estimator, which works well on nonstationary time series. Finally, lots of experiments validate that HCE algorithm achieves state-of-the-art accuracy when time series are nonstationary, compared to the other baseline algorithms.
