Shapley-PC: constraint-based causal structure learning with a Shapley inspired framework
File(s)
Accepted version
Author(s)
Russo, Fabrizio
Toni, francesca
Type
Conference Paper
Abstract
Causal Structure Learning (CSL), also referred to as causal discovery, amounts to extracting causal relations among variables in data. CSL enables the estimation of causal effects from observational data alone, avoiding the need to perform real life experiments. Constraint-based CSL leverages conditional independence tests to perform causal discovery. We propose Shapley-PC, a novel method to improve constraint-based CSL algorithms by using Shapley values over the possible conditioning sets, to decide which variables are responsible for the observed conditional (in)dependences. We prove soundness, completeness and asymptotic consistency of Shapley-PC and run a simulation
study showing that our proposed algorithm is superior to existing versions of PC.
study showing that our proposed algorithm is superior to existing versions of PC.
Date Issued
2025-05-09
Date Acceptance
2025-01-27
Citation
Proceedings of the Fourth Conference on Causal Learning and Reasoning, 2025, 275, pp.292-339
ISSN
2640-3498
Publisher
PMLR
Start Page
292
End Page
339
Journal / Book Title
Proceedings of the Fourth Conference on Causal Learning and Reasoning
Volume
275
Copyright Statement
© 2025 F. Russo & F. Toni. This is the author’s accepted manuscript made available under a CC-BY licence in accordance with Imperial’s Research Publications Open Access policy (www.imperial.ac.uk/oa-policy)
License URL
Source
4th Conference on Causal Learning and Reasoning (CLeaR 2025)
Publication Status
Published
Start Date
2025-05-07
Finish Date
2025-05-09
Coverage Spatial
Lausanne, Switzerland
Date Publish Online
2025-05-09