Gradient-Based Fine-Tuning of Expert Bayesian Network Classifiers: Application in Forensic Porcine Wound Age Assessment

Joseph Mietkiewicz, Anders L Madsen, Thomas Dyhre Nielsen, Cecilie Bækgård, Henrik Elvang Jensen, Kristiane Barington
Proceedings of the 13th International Conference on Probabilistic Graphical Models, PMLR 343:629-645, 2026.

Abstract

Bayesian network classifiers built from expert knowledge encode valuable domain understanding but may be improved when labelled data become available. We propose \textsc{EB-GFTBN}, a Gradient-based Fine-Tuning algorithm for Bayesian Networks with Expert-Bounded parameter updates. Building on the fine-tuning framework of El Hindi (2014) and the Brier-score variant of Mietkiewicz and Madsen (2022), we derive natural-gradient updates from the derivative of the Brier score and extend the algorithm to mixed discrete–continuous Bayesian networks. To remain faithful to the expert specification, each parameter is constrained to a relative tolerance around its expert elicited value. The methodology is evaluated in a case study regarding forensic age assessment of porcine wounds, where the parameters of a Bayesian network classifier were elicited from veterinary experts. Across $100$ cross-validation folds, \textsc{EB-GFTBN} improves both accuracy and Brier score over the expert baseline and the penalised EM alternative, while keeping every parameter inside the expert-allowed window.

Cite this Paper

BibTeX
@InProceedings{pmlr-v343-mietkiewicz26a, title = {Gradient-Based Fine-Tuning of Expert {Bayesian} Network Classifiers: Application in Forensic Porcine Wound Age Assessment}, author = {Mietkiewicz, Joseph and Madsen, Anders L and Nielsen, Thomas Dyhre and B{\ae}kg{\aa}rd, Cecilie and Jensen, Henrik Elvang and Barington, Kristiane}, booktitle = {Proceedings of the 13th International Conference on Probabilistic Graphical Models}, pages = {629--645}, year = {2026}, editor = {Camps-Valls, Gustau and Leonelli, Manuele and Varando, Gherardo}, volume = {343}, series = {Proceedings of Machine Learning Research}, month = {09--11 Sep}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v343/main/assets/mietkiewicz26a/mietkiewicz26a.pdf}, url = {https://proceedings.mlr.press/v343/mietkiewicz26a.html}, abstract = {Bayesian network classifiers built from expert knowledge encode valuable domain understanding but may be improved when labelled data become available. We propose \textsc{EB-GFTBN}, a Gradient-based Fine-Tuning algorithm for Bayesian Networks with Expert-Bounded parameter updates. Building on the fine-tuning framework of El Hindi (2014) and the Brier-score variant of Mietkiewicz and Madsen (2022), we derive natural-gradient updates from the derivative of the Brier score and extend the algorithm to mixed discrete–continuous Bayesian networks. To remain faithful to the expert specification, each parameter is constrained to a relative tolerance around its expert elicited value. The methodology is evaluated in a case study regarding forensic age assessment of porcine wounds, where the parameters of a Bayesian network classifier were elicited from veterinary experts. Across $100$ cross-validation folds, \textsc{EB-GFTBN} improves both accuracy and Brier score over the expert baseline and the penalised EM alternative, while keeping every parameter inside the expert-allowed window.} }
Endnote
%0 Conference Paper %T Gradient-Based Fine-Tuning of Expert Bayesian Network Classifiers: Application in Forensic Porcine Wound Age Assessment %A Joseph Mietkiewicz %A Anders L Madsen %A Thomas Dyhre Nielsen %A Cecilie Bækgård %A Henrik Elvang Jensen %A Kristiane Barington %B Proceedings of the 13th International Conference on Probabilistic Graphical Models %C Proceedings of Machine Learning Research %D 2026 %E Gustau Camps-Valls %E Manuele Leonelli %E Gherardo Varando %F pmlr-v343-mietkiewicz26a %I PMLR %P 629--645 %U https://proceedings.mlr.press/v343/mietkiewicz26a.html %V 343 %X Bayesian network classifiers built from expert knowledge encode valuable domain understanding but may be improved when labelled data become available. We propose \textsc{EB-GFTBN}, a Gradient-based Fine-Tuning algorithm for Bayesian Networks with Expert-Bounded parameter updates. Building on the fine-tuning framework of El Hindi (2014) and the Brier-score variant of Mietkiewicz and Madsen (2022), we derive natural-gradient updates from the derivative of the Brier score and extend the algorithm to mixed discrete–continuous Bayesian networks. To remain faithful to the expert specification, each parameter is constrained to a relative tolerance around its expert elicited value. The methodology is evaluated in a case study regarding forensic age assessment of porcine wounds, where the parameters of a Bayesian network classifier were elicited from veterinary experts. Across $100$ cross-validation folds, \textsc{EB-GFTBN} improves both accuracy and Brier score over the expert baseline and the penalised EM alternative, while keeping every parameter inside the expert-allowed window.
APA
Mietkiewicz, J., Madsen, A.L., Nielsen, T.D., Bækgård, C., Jensen, H.E. & Barington, K.. (2026). Gradient-Based Fine-Tuning of Expert Bayesian Network Classifiers: Application in Forensic Porcine Wound Age Assessment. Proceedings of the 13th International Conference on Probabilistic Graphical Models, in Proceedings of Machine Learning Research 343:629-645 Available from https://proceedings.mlr.press/v343/mietkiewicz26a.html.

Related Material