ENCODE: EXPLAINABILITY OF PREDICTIVE UNCERTAINTY MODELS UNDER DRIFT IN THE TELECOM DOMAIN

Machine learning (ML) models deployed in dynamic environments are prone to distributional shifts that degrade predictive reliability. While uncertainty quantification and drift detection are widely studied, their interaction with explainability remains insufficiently understood, particularly for regression tasks under covariate shift. This paper presents a unified experimental framework that integrates uncer- tainty quantification (UQ), calibration, drift analysis, and explainable AI (XAI) for regression models in a real-world telecommunications setting. We evaluate Bayesian neural networks (BNNs) with Monte Carlo Dropout and deep ensembles, quantifying predictive uncertainty using variance and assessing calibration quality via Expected Normalized Calibration Error (ENCE). The results show that BNNs exhibit higher sensitiv- ity to drift through pronounced increases in predictive uncertainty, while ensemble models provide more stable but less adaptive estimates. Importantly, explanation patterns consistently track uncertainty degradation, highlighting the value of XAI as a diagnostic tool for drift-aware model monitoring and lifecycle management in non-stationary environments.