[2603.25670] Uncertainty-Guided Label Rebalancing for CPS Safety Monitoring

Computer Science > Machine Learning arXiv:2603.25670 (cs) [Submitted on 26 Mar 2026] Title:Uncertainty-Guided Label Rebalancing for CPS Safety Monitoring Authors:John Ayotunde, Qinghua Xu, Guancheng Wang, Lionel C. Briand View a PDF of the paper titled Uncertainty-Guided Label Rebalancing for CPS Safety Monitoring, by John Ayotunde and 3 other authors View PDF HTML (experimental) Abstract:Safety monitoring is essential for Cyber-Physical Systems (CPSs). However, unsafe events are rare in real-world CPS operations, creating an extreme class imbalance that degrades safety predictors. Standard rebalancing techniques perform poorly on time-series CPS telemetry, either generating unrealistic synthetic samples or overfitting on the minority class. Meanwhile, behavioral uncertainty in CPS operations, defined as the degree of doubt or uncertainty in CPS decisions , is often correlated with safety outcomes but unexplored in safety monitoring. To that end, we propose U-Balance, a supervised approach that leverages behavioral uncertainty to rebalance imbalanced datasets prior to training a safety predictor. U-Balance first trains a GatedMLP-based uncertainty predictor that summarizes each telemetry window into distributional kinematic features and outputs an uncertainty score. It then applies an uncertainty-guided label rebalancing (uLNR) mechanism that probabilistically relabels \textit{safe}-labeled windows with unusually high uncertainty as \textit{unsafe}, thereby enriching the m...