Problem
Research questions and friction points this paper is trying to address.
Time-series anomaly detection
Temporal change disambiguation
Distribution drift
Non-stationarity
MORA: Modeling Observed Changes for Drift-Robust Time-Series Anomaly Detection
🤖 AI Summary
This study addresses the challenge of distinguishing distribution shifts from genuine anomalies in non-stationary time series, which often degrades detection performance. To tackle this issue, we propose a temporal variation disambiguation framework that quantifies contextual support through reconstruction discrepancies between long- and short-term views. Furthermore, it employs a data-dependent mechanism to conservatively adjust local anomaly scores, suppressing shift-induced interference only when sufficient evidence is available. By integrating multi-scale modeling with adaptive score correction, this unsupervised approach demonstrates strong robustness against distribution drift while maintaining high sensitivity to anomalies across four benchmark datasets.
📝 Abstract
Time-series anomaly detection (TSAD) identifies deviations from patterns learned from historical data. In non-stationary settings, distribution drift and true anomalies can cause similar local changes, making it difficult to tell whether a deviation reflects abnormality or evolving context. Existing methods typically adapt to detected shifts or learn drift-insensitive representations, but do not resolve this ambiguity. We define this problem as \emph{temporal change disambiguation}: determining whether a local deviation is explained by broader temporal evolution. We introduce MORA, a drift-robust TSAD framework that reconstructs the same local target from paired short- and long-term views. The reconstruction gap measures contextual support for a local deviation, and a data-dependent correction mechanism conservatively adjusts the primary local anomaly score. Context can only reduce the score when it improves reconstruction of the same target. MORA needs neither drift annotations nor online adaptation. Experiments on four TSAD benchmarks show strong robustness to non-stationarity while preserving sensitivity to genuine anomalies.
Innovation
Methods, ideas, or system contributions that make the work stand out.
Time-series anomaly detection
Temporal change disambiguation
Drift-robustness
Multi-view reconstruction
Non-stationarity
🔎 Similar Papers
No similar papers found.
