Fast Flow Matching based Conditional Independence Tests for Causal Discovery

2026-02-09

AI Summary

Constraint-based causal discovery is computationally expensive due to its reliance on a large number of conditional independence (CI) tests. This work proposes FMCIT, a flow-matching–based CI testing method that, for the first time, introduces flow matching into CI testing, enabling efficient reuse of a single trained model across all CI queries. Furthermore, the authors develop GPC-FMCIT, a two-stage guided PC skeleton learning framework integrated with a budget control strategy that explicitly limits the number of CI tests while preserving high statistical power. Experimental results demonstrate that FMCIT effectively controls Type I error rates, and GPC-FMCIT achieves a superior trade-off between accuracy and efficiency compared to existing methods on both synthetic and real-world datasets.

Technology Category

• Machine Learning: Causal Learning
• Reasoning under Uncertainty: Causality
• Constraint Satisfaction and Optimization: Constraint Learning and Acquisition

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: Robustness and generalizability of Web mining methods
• Security and Privacy: Large-scale security measurements

Abstract

Constraint-based causal discovery methods require a large number of conditional independence (CI) tests, which severely limits their practical applicability due to high computational complexity. Therefore, it is crucial to design an algorithm that accelerates each individual test. To this end, we propose the Flow Matching-based Conditional Independence Test (FMCIT). The proposed test leverages the high computational efficiency of flow matching and requires the model to be trained only once throughout the entire causal discovery procedure, substantially accelerating causal discovery. According to numerical experiments, FMCIT effectively controls type-I error and maintains high testing power under the alternative hypothesis, even in the presence of high-dimensional conditioning sets. In addition, we further integrate FMCIT into a two-stage guided PC skeleton learning framework, termed GPC-FMCIT, which combines fast screening with guided, budgeted refinement using FMCIT. This design yields explicit bounds on the number of CI queries while maintaining high statistical power. Experiments on synthetic and real-world causal discovery tasks demonstrate favorable accuracy-efficiency trade-offs over existing CI testing methods and PC variants.

Problem

• conditional independence test
• causal discovery
• computational complexity
• constraint-based methods
• high-dimensional conditioning

Innovation

• Flow Matching
• Conditional Independence Test
• Causal Discovery
• Efficient Inference
• PC Algorithm