Uncertainty-Guided Label Rebalancing (uLNR)

Auto-generated stub. Edit this file to add more details.

A mechanism that probabilistically relabels safe-labeled time-series windows with high behavioral uncertainty as unsafe to augment the minority class for safety monitoring.

Why It Matters

It is the core novel technique used to address extreme class imbalance by re-labeling uncertain ‘safe’ examples as ‘unsafe’ to enrich the minority class.

Evidence

applies an uncertainty-guided label rebalancing (uLNR) mechanism that probabilistically relabels \textit{safe}-labeled windows with unusually high uncertainty as \textit{unsafe}

Related Papers

• openalex-2603.25670-uncertainty-guided-label-rebalancing-for-cps-safety-monitori