Saved in:
Bibliographic Details
Main Authors: Fehri, Ghita Fassy El, Bellet, Aurélien, Bastien, Philippe
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2512.01708
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913020385427456
author Fehri, Ghita Fassy El
Bellet, Aurélien
Bastien, Philippe
author_facet Fehri, Ghita Fassy El
Bellet, Aurélien
Bastien, Philippe
contents Learning the structure of a Bayesian network from decentralized data poses two major challenges: (i) ensuring rigorous privacy guarantees for participants, and (ii) avoiding communication costs that scale poorly with dimensionality. In this work, we introduce Fed-Sparse-BNSL, a novel federated method for learning linear Gaussian Bayesian network structures that addresses both challenges. By combining differential privacy with greedy updates that target only a few relevant edges per participant, Fed-Sparse-BNSL efficiently uses the privacy budget while keeping communication costs low. Our careful algorithmic design preserves model identifiability and enables accurate structure estimation. Experiments on synthetic and real datasets demonstrate that Fed-Sparse-BNSL achieves utility close to non-private baselines while offering substantially stronger privacy and communication efficiency.
format Preprint
id arxiv_https___arxiv_org_abs_2512_01708
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Differentially Private and Federated Structure Learning in Bayesian Networks
Fehri, Ghita Fassy El
Bellet, Aurélien
Bastien, Philippe
Machine Learning
Learning the structure of a Bayesian network from decentralized data poses two major challenges: (i) ensuring rigorous privacy guarantees for participants, and (ii) avoiding communication costs that scale poorly with dimensionality. In this work, we introduce Fed-Sparse-BNSL, a novel federated method for learning linear Gaussian Bayesian network structures that addresses both challenges. By combining differential privacy with greedy updates that target only a few relevant edges per participant, Fed-Sparse-BNSL efficiently uses the privacy budget while keeping communication costs low. Our careful algorithmic design preserves model identifiability and enables accurate structure estimation. Experiments on synthetic and real datasets demonstrate that Fed-Sparse-BNSL achieves utility close to non-private baselines while offering substantially stronger privacy and communication efficiency.
title Differentially Private and Federated Structure Learning in Bayesian Networks
topic Machine Learning
url https://arxiv.org/abs/2512.01708