cs.LGNov 6, 2026

Multi-objective drift-aware MLOps agent for model retraining

Authors: Kumuda Valli, Anli Ji

Organizations: Computer Science California State University, Fullerton Fullerton, United States of America

Abstract

Deployed ML models face non-stationary data and delayed labels, yet most retraining strategies rely on fixed schedules or brittle drift thresholds that ignore operational cost. In this setting, retraining is a sequential decision problem rather than a one-time threshold choice. We present DARE (Drift-Aware Retraining Engine), an offline, event-driven retraining framework that separates ingestion, monitoring, policy decisions, training, and logging through explicit schemas and deterministic replay. We compare DARE against practical heuristic retraining baselines, including fixed-interval, threshold-persistence, and action-rate-matched fixed scheduling policies. On Airline, the locked LinUCB policy achieves a higher mean F1 while substantially reducing intervention frequency relative to heuristic and action-rate-matched baselines. Additional runtime-calibrated evaluation shows that these gains persist when measured using the cumulative update wall-clock time. These results suggest that conservative sequential retraining policies can improve performance while reducing unnecessary operational overhead under delayed feedback.

Figures & tables

Explore similar work

CardsList
  1. Retrain, Alert, or Rollback? A Three-Response Decision Framework for Production ML Under Concept Drift

    Nov 1, 2026Ali Sadeghian, Alireza Shafaei, Hamid Hadibeik +1Concept DriftDrifting Models

  2. ADORN: Adaptive Drift handling for Open RAN using Reinforcement Learning

    Jul 9, 2026Ashit Kumar Subudhi, Bhargav Chirumamilla, Shubham Vaishnav +5Concept DriftOpen Radio Access Network