SEP 9, 2026 · PREPRINT
Evaluating Model Retraining under Drift: Paired Comparisons of Cumulative Subgroup Disparity
arXiv
This is an unrefereed methodological study on classifier retraining policies using simulation and observational data, presenting exploratory analyses without peer review.
Preprint: Not Peer ReviewedAaron Ceross
Reported
Mean cumulative disparity reducti…0.04 to 0.88 percentage points
Agreement between finite-window a…69 to 92 percent of trajectories
Sample size per condition (simula…400