RED-F: Reconstruction-Elimination based Dual-stream Contrastive Forecasting for Multivariate Time Series Anomaly Prediction

TL;DR

RED-F introduces a dual-stream contrastive forecasting framework that improves multivariate time series anomaly prediction by constructing normal pattern baselines and comparing future predictions to detect anomalies more effectively.

Contribution

The paper proposes a novel RED-F framework combining reconstruction-elimination and contrastive forecasting to enhance anomaly prediction accuracy in multivariate time series.

Findings

RED-F outperforms existing methods on multiple datasets

The contrastive approach improves anomaly detection robustness

Multi-Series Prediction training enhances sensitivity to anomalies

Abstract

Anomaly prediction (AP) in multivariate time series (MTS) is crucial to ensure system dependability. Existing methods either focus solely on whether an anomaly is imminent without providing precise predictions for the future anomaly, or performing predictions directly on historical data, which is easily drowned out by the normal patterns. To address the challenges in AP task, we propose RED-F, a novel framework comprised of the Reconstruction-Elimination Model (REM) and the Dual-stream Contrastive Forecasting Model (DFM). We utilize REM to construct a baseline of normal patterns from historical data, providing a foundation for subsequent predictions of anomalies. Then DFM simultaneously predicts both the constructed normal pattern and the current window, employing a contrastive forecast that transforms the difficult AP task into a simpler, more robust task of relative trajectory comparison by computing the divergence between these two predictions. To enable the forecasting model to generate a prediction not easily obscured by normal patterns, we propose a Multi-Series Prediction (MSP) training objective to enhance its sensitivity to the current window. Extensive experiments on multiple real-world datasets demonstrate the superior capability of RED-F in anomaly prediction tasks. Our code is available at http://github.com/PenyChen/RED-F.