A Characterization of Markov Equivalence Classes of Relational Causal Models under Path Semantics

Sanghack Lee Penn State University, Vasant Honavar Penn State University
Proceedings of the 32nd Conference on Uncertainty in Artificial Intelligence, PMLR R14:128-137, 2016.

Abstract

Relational Causal Models (RCM) generalize Causal Bayesian Networks so as to extend causal discovery to relational domains. We provide a novel and elegant characterization of the Markov equivalence of RCMs under path semantics. We introduce a novel representation of unshielded triples that allows us to efficiently determine whether an RCM is Markov equivalent to another. Under path semantics, we provide a sound and complete algorithm for recovering the structure of an RCM from conditional independence queries. Our analysis also suggests ways to improve the orientation recall of algorithms for learning the structure of RCM under bridge burning semantics as well.

Cite this Paper

BibTeX
@InProceedings{pmlr-vR14-university16d, title = {A Characterization of {M}arkov Equivalence Classes of Relational Causal Models under Path Semantics}, author = {University, Sanghack Lee Penn State and University, Vasant Honavar Penn State}, booktitle = {Proceedings of the 32nd Conference on Uncertainty in Artificial Intelligence}, pages = {128--137}, year = {2016}, editor = {Ihler, Alexander and Janzing, Dominik}, volume = {R14}, series = {Proceedings of Machine Learning Research}, month = {25--29 Jun}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/r14/main/assets/university16d/university16d.pdf}, url = {https://proceedings.mlr.press/r14/university16d.html}, abstract = {Relational Causal Models (RCM) generalize Causal Bayesian Networks so as to extend causal discovery to relational domains. We provide a novel and elegant characterization of the Markov equivalence of RCMs under path semantics. We introduce a novel representation of unshielded triples that allows us to efficiently determine whether an RCM is Markov equivalent to another. Under path semantics, we provide a sound and complete algorithm for recovering the structure of an RCM from conditional independence queries. Our analysis also suggests ways to improve the orientation recall of algorithms for learning the structure of RCM under bridge burning semantics as well.}, note = {Reissued by PMLR on 04 October 2026.} }
Endnote
%0 Conference Paper %T A Characterization of Markov Equivalence Classes of Relational Causal Models under Path Semantics %A Sanghack Lee Penn State University %A Vasant Honavar Penn State University %B Proceedings of the 32nd Conference on Uncertainty in Artificial Intelligence %C Proceedings of Machine Learning Research %D 2016 %E Alexander Ihler %E Dominik Janzing %F pmlr-vR14-university16d %I PMLR %P 128--137 %U https://proceedings.mlr.press/r14/university16d.html %V R14 %X Relational Causal Models (RCM) generalize Causal Bayesian Networks so as to extend causal discovery to relational domains. We provide a novel and elegant characterization of the Markov equivalence of RCMs under path semantics. We introduce a novel representation of unshielded triples that allows us to efficiently determine whether an RCM is Markov equivalent to another. Under path semantics, we provide a sound and complete algorithm for recovering the structure of an RCM from conditional independence queries. Our analysis also suggests ways to improve the orientation recall of algorithms for learning the structure of RCM under bridge burning semantics as well. %Z Reissued by PMLR on 04 October 2026.
APA
University, S.L.P.S. & University, V.H.P.S.. (2016). A Characterization of Markov Equivalence Classes of Relational Causal Models under Path Semantics. Proceedings of the 32nd Conference on Uncertainty in Artificial Intelligence, in Proceedings of Machine Learning Research R14:128-137 Available from https://proceedings.mlr.press/r14/university16d.html. Reissued by PMLR on 04 October 2026.

Related Material