Bayesian Network Structural Consensus via Greedy Min-Cut Analysis
Proceedings of the AAAI Conference on Artificial Intelligence · 14.03.2026 · 40 (43) · 36749-36756
DOI: 10.1609/aaai.v40i43.41000 ↗Kurz gefasst
Fachartikel in der Zeitschrift Proceedings of the AAAI Conference on Artificial Intelligence, online erschienen am 14.03.2026. Verfasst von Pablo Torrijos und 3 weiteren Autor:innen.
Abstract (Originalsprache)
This paper presents the Min-Cut Bayesian Network Consensus (MCBNC) algorithm, a greedy method for structural consensus of Bayesian Networks (BNs), with applications in federated learning and model aggregation. MCBNC prunes weak edges from an initial unrestricted fusion using a structural score based on min-cut analysis, integrated into a modified Backward Equivalence Search (BES) phase of the Greedy Equivalence Search (GES) algorithm. The score quantifies edge support across input networks and is computed using max-flow. Unlike methods with fixed treewidth bounds, MCBNC introduces a pruning threshold θ that can be selected post hoc using only structural information. Experiments on real-world BNs show that MCBNC yields sparser, more accurate consensus structures than both canonical fusion and the input networks. The method is scalable, data-agnostic, and well-suited for distributed or federated structural learning of BNs or causal discovery.
Zitieren
APA
Torrijos, P., Puerta, J. M., Aledo, J. A., & Gámez, J. A. (2026). Bayesian Network Structural Consensus via Greedy Min-Cut Analysis. Proceedings of the AAAI Conference on Artificial Intelligence, 40(43), 36749-36756. https://doi.org/10.1609/aaai.v40i43.41000
Vancouver
Torrijos P, Puerta JM, Aledo JA, Gámez JA. Bayesian Network Structural Consensus via Greedy Min-Cut Analysis. Proceedings of the AAAI Conference on Artificial Intelligence. 2026;40(43):36749-36756. doi:10.1609/aaai.v40i43.41000
BibTeX
@article{Torrijos2026,
author = {Torrijos, Pablo and Puerta, Jose M. and Aledo, Juan A. and Gámez, José A.},
title = {Bayesian Network Structural Consensus via Greedy Min-Cut Analysis},
journal = {Proceedings of the AAAI Conference on Artificial Intelligence},
year = {2026},
volume = {40},
number = {43},
pages = {36749--36756},
doi = {10.1609/aaai.v40i43.41000}
}
EUVIMED