Parallel Adaptive Multi-Objective Evolutionary Learning of Discretized Bayesian Network Classifiers for Clinical Data

2026-05-27

AI Summary

This work addresses the limitations of Bayesian network classifiers in clinical settings—namely, low learning efficiency, weak generalization, and insufficient interpretability—by proposing Baymex, a parallel learning framework based on multi-objective evolutionary algorithms. Baymex jointly optimizes cross-entropy loss and the Bayesian Information Criterion (BIC) complexity term, while incorporating an adaptive regularization mechanism to mitigate overfitting. The approach significantly enhances training efficiency, achieving over 54× speedup on a 16-core CPU, without compromising model compactness or clinical interpretability. Evaluated on three real-world clinical datasets, Baymex matches or surpasses the predictive performance of established models, thereby providing the first empirical validation of the effectiveness and scalability of multi-objective Bayesian network classifiers in clinical applications.

Abstract

Bayesian Networks (BNs) are of interest from an explainable AI viewpoint, offering transparent probabilistic models for decision support. Baymex is a recently introduced multi-objective evolutionary algorithm for learning discretized BNs, enabling experts to trade-off different objectives of interest, such as likelihood, model complexity, and prior beliefs. While Baymex has been shown to outperform state-of-the-art BN learning approaches, Baymex still 1) requires a lot of computation time and 2) has only been evaluated on synthetic data. To improve scalability, we introduce a parallelization strategy as well as a mechanism that enables adaptively steering optimization toward networks that overfit less. We furthermore reconfigure Baymex to train a BN classifier through multi-objective optimization of cross-entropy loss and the BIC complexity term so as to evaluate its performance on real-world clinical classification tasks. Besides observing speedups up to over 54 times on a 16-core CPU, comparisons against clinically familiar baselines (decision trees, logistic regression, naive Bayes, and random forests) on two open-source (RADCURE and SUPPORT) and one in-house dataset, show that Baymex obtains statistically similar or better predictive performance while producing compact, clinically inspectable BNs. Importantly, Baymex finds multiple plausible BN classifiers that contain predictors consistent with established clinical factors.

Problem

• Bayesian Networks
• Multi-Objective Evolutionary Algorithm
• Clinical Data
• Scalability
• Explainable AI

Innovation

• parallelization
• adaptive optimization
• multi-objective evolutionary algorithm
• discretized Bayesian network
• clinical interpretability