Scalable causal structure learning via amortized conditional independence testing Conference Paper

  


Authors: Leiner, J.; Manzo, B.; Ramdas, A.; Tansey, W.
Title: Scalable causal structure learning via amortized conditional independence testing
Conference Title: 4th Conference on Causal Learning and Reasoning (CLeaR 2025)
Abstract: Controlling false positives (Type I errors) through statistical hypothesis testing is a foundation of modern scientific data analysis. Existing causal structure discovery algorithms either do not provide Type I error control or cannot scale to the size of modern scientific datasets. We consider a variant of the causal discovery problem with two sets of nodes, where the only edges of interest form a bipartite causal subgraph between the sets. We develop Scalable Causal Structure Learning (SCSL), a method for causal structure discovery on bipartite subgraphs that provides Type I error control. SCSL recasts the discovery problem as a simultaneous hypothesis testing problem and uses discrete optimization over the set of possible confounders to obtain an upper bound on the test statistic for each edge. Semi-synthetic simulations demonstrate that SCSL scales to handle graphs with hundreds of nodes while maintaining error control and good power. We demonstrate the practical applicability of the method by applying it to a cancer dataset to reveal connections between somatic gene mutations and metastases to different tissues. © 2025 Elsevier B.V., All rights reserved.
Keywords: optimization; statistical tests; hypothesis testing; false positive; type-i error; errors; learning systems; causal inference; directed acyclic graph; conditional independences; directed acyclic graphs; acyclic graphs; causal inferences; causal structure learning; error control; statistical hypothesis testing
Journal Title Proceedings of Machine Learning Research
Volume: 275
Conference Dates: 2025 May 7-9
Conference Location: Lausanne, Switzerland
ISBN: 2640-3498
Publisher: Journal Machine Learning Research  
Date Published: 2025-01-01
Start Page: 174
End Page: 200
Language: English
PROVIDER: scopus
DOI/URL:

https://www.scopus.com/inward/record.uri?eid=2-s2.0-105014181422&partnerID=40&md5=dc410200377f2cf0e5d130a362161d5a
Notes: Conference Paper -- Source: Scopus