Train, Mutate, or Reward? A Unified View of Supervised Ensembling for Time Series Anomaly Detection.

ICLR 2026 Conference Submission19679 Authors

Keywords: Time Series ; Anomaly Detection ; Supervised Ensembling

TL;DR: We propose and rigorously evaluate three supervised ensembling strategies : classical ML, reinforcement learning, and genetic programming, as strong baselines for time series anomaly detection, outperforming individual detectors and model selection.

Abstract:

Time series anomaly detection (TSAD) is a long-standing and extensively studied problem with applications across a large panel of domains. Despite the maturity of the field, recent benchmark studies have revealed that no single detection method consistently outperforms others across diverse datasets. While model selection approaches (i.e., choosing the best detector for a given scenario) have shown promising results, their effectiveness remains inherently limited by the performance ceiling of existing individual detectors.
To address this limitation, supervised ensembling offers a promising path to surpass individual detectors by learning to combine their strengths. In this work, we unify and formalize the problem of supervised ensemble-based anomaly detection in time series, and introduce three principled strategies for learning such ensembles: (1) classical Machine Learning, (2) Reinforcement Learning, and (3) Genetic Programming. We perform a rigorous comparative evaluation across these strategies using identical model components, inputs, and experimental conditions to ensure fairness. Our findings not only highlight the strengths and trade-offs of each approach, but also illuminate promising directions, paving the road for future research on this topic.

Primary Area: learning on time series and dynamical systems

Submission Number: 19679