Efficient Ensemble Conditional Independence Test Framework for Causal Discovery

Researchers introduce E-CIT (Ensemble Conditional Independence Test), a new framework that significantly reduces computational costs in causal discovery by partitioning data into subsets and aggregating results. The method achieves linear computational complexity while maintaining competitive performance, particularly on real-world datasets.

• →E-CIT framework reduces computational complexity of causal discovery from high-order to linear with respect to sample size.
• →The method uses a divide-and-aggregate strategy with novel p-value combination grounded in stable distributions.
• →Framework offers theoretical consistency guarantees under mild conditions on subtests.
• →Experimental results show significant reduction in computational burden while maintaining competitive performance.
• →The approach demonstrates particular improvement in complex testing scenarios on real-world datasets.