CUTS+: High-dimensional Causal Discovery from Irregular Time-series
This addresses the challenge of scalable causal reasoning for complex, high-dimensional time-series data, representing an incremental advancement over prior Granger-causality-based approaches.
The paper tackles the problem of causal discovery in high-dimensional, irregularly sampled time-series data, where existing methods degrade due to redundant network designs and missing entries. The proposed CUTS+ method significantly improves performance, as demonstrated on simulated, quasi-real, and real datasets.
Causal discovery in time-series is a fundamental problem in the machine learning community, enabling causal reasoning and decision-making in complex scenarios. Recently, researchers successfully discover causality by combining neural networks with Granger causality, but their performances degrade largely when encountering high-dimensional data because of the highly redundant network design and huge causal graphs. Moreover, the missing entries in the observations further hamper the causal structural learning. To overcome these limitations, We propose CUTS+, which is built on the Granger-causality-based causal discovery method CUTS and raises the scalability by introducing a technique called Coarse-to-fine-discovery (C2FD) and leveraging a message-passing-based graph neural network (MPGNN). Compared to previous methods on simulated, quasi-real, and real datasets, we show that CUTS+ largely improves the causal discovery performance on high-dimensional data with different types of irregular sampling.