Forecasting Anomaly Precursors via Uncertainty-Aware Time-Series Ensembles

TL;DR

FATE is an unsupervised ensemble-based framework that predicts anomalies early in time-series data by quantifying uncertainty, outperforming existing methods without needing labeled anomalies.

Contribution

The paper introduces FATE, a novel ensemble-based approach that detects anomaly precursors proactively using uncertainty quantification and a new evaluation metric, PTaPR.

Findings

FATE improves early warning detection metrics significantly over baselines.

It achieves an average of 19.9% PTaPR AUC improvement.

FATE requires no labeled anomaly data during training.

Abstract

Detecting anomalies in time-series data is critical in domains such as industrial operations, finance, and cybersecurity, where early identification of abnormal patterns is essential for ensuring system reliability and enabling preventive maintenance. However, most existing methods are reactive: they detect anomalies only after they occur and lack the capability to provide proactive early warning signals. In this paper, we propose FATE (Forecasting Anomalies with Time-series Ensembles), a novel unsupervised framework for detecting Precursors-of-Anomaly (PoA) by quantifying predictive uncertainty from a diverse ensemble of time-series forecasting models. Unlike prior approaches that rely on reconstruction errors or require ground-truth labels, FATE anticipates future values and leverages ensemble disagreement to signal early signs of potential anomalies without access to target values at inference time. To rigorously evaluate PoA detection, we introduce Precursor Time-series Aware Precision and Recall (PTaPR), a new metric that extends the traditional Time-series Aware Precision and Recall (TaPR) by jointly assessing segment-level accuracy, within-segment coverage, and temporal promptness of early predictions. This enables a more holistic assessment of early warning capabilities that existing metrics overlook. Experiments on five real-world benchmark datasets show that FATE achieves an average improvement of 19.9 percentage points in PTaPR AUC and 20.02 percentage points in early detection F1 score, outperforming baselines while requiring no anomaly labels. These results demonstrate the effectiveness and practicality of FATE for real-time unsupervised early warning in complex time-series environments.