AA-Forecast: Anomaly-Aware Forecast for Extreme Events

TL;DR

This paper introduces AA-Forecast, an anomaly-aware time series forecasting framework that automatically detects anomalies and leverages them to improve prediction accuracy and reduce uncertainty during extreme events.

Contribution

It presents a novel framework that automatically extracts anomalies and uses an attention mechanism to enhance forecasting accuracy during extreme events.

Findings

Consistently outperforms existing models on multiple datasets.

Reduces forecast uncertainty in the presence of anomalies.

Effectively leverages anomaly effects for improved predictions.

Abstract

Time series models often deal with extreme events and anomalies, both prevalent in real-world datasets. Such models often need to provide careful probabilistic forecasting, which is vital in risk management for extreme events such as hurricanes and pandemics. However, it is challenging to automatically detect and learn to use extreme events and anomalies for large-scale datasets, which often require manual effort. Hence, we propose an anomaly-aware forecast framework that leverages the previously seen effects of anomalies to improve its prediction accuracy during and after the presence of extreme events. Specifically, the framework automatically extracts anomalies and incorporates them through an attention mechanism to increase its accuracy for future extreme events. Moreover, the framework employs a dynamic uncertainty optimization algorithm that reduces the uncertainty of forecasts in an online manner. The proposed framework demonstrated consistent superior accuracy with less uncertainty on three datasets with different varieties of anomalies over the current prediction models.