Forecasting Anomaly Precursors via Uncertainty-Aware Time-Series Ensembles

TL;DR

FATE (Forecasting Anomalies with Time-series Ensembles 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, is proposed.

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.

Figures (8)

• Figure 1: Comparison between traditional anomaly detection and Precursor-of-Anomaly (PoA) detection. (a) Traditional methods detect anomalies after their occurrence, limiting the ability to prevent failures. (b) PoA detection identifies early warning signals prior to anomaly onset, enabling proactive intervention.
• Figure 2: Overview of the proposed FATE framework. The input sequence is fed into an ensemble of forecasting models to predict future time steps. Uncertainty-based precursor detection is performed using the ensemble outputs, and early detection performance is quantitatively evaluated using the PTaPR metric.
• Figure 3: Example of predicting anomaly precursors at future time steps based on input at time $t$. The third window ($w=2$) forecasts $L_y=5$ future steps from current time $t+L_x-1=11$, and a precursor is detected at time step 13.
• Figure 4: Illustration of early detection, partial detection, and ambiguous instances in the PTaPR metric. Anomalies and predictions are represented as instance-level segments along the x-axis. For example, anomaly $a_1$ spans indices 1 to 6, while early prediction $p'_3$ detects an anomaly between indices 13 and 16.
• Figure 5: Example of PTaPR computation for anomaly segments ($a_1, a_2$) and prediction segments ($p_1, p_2$). The overlap score $O(a, p, p')$ is the sum of detected anomalies $|a \cap p|$, early predictions $|a \cap p'|$, and the ambiguous instance score $S(a', p)$. For $a_1$, the overlap score is 3 with an early detection reward of 0.29. For $a_2$, the overlap score is approximately 5.88, reflecting the contribution of the sigmoid-weighted ambiguous instances.