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

2025-11-25

AI Summary

Weak and easily masked by normal patterns, anomaly precursors (APs) in multivariate time series (MTS) pose significant challenges for unsupervised detection. Existing methods—relying solely on modeling normal data—suffer from reconstruction dominance, leading to insufficient sensitivity for early AP identification. To address this, we propose RED-F: (1) a Reconstruction Elimination Mechanism (REM) that isolates clean normal baselines by suppressing interference; (2) a Dual-Flow Contrastive Prediction Model (DFM) that reframes AP detection as trajectory discrepancy analysis between a normal flow and a perturbed flow; and (3) a Multi-Sequence Prediction (MSP) objective that explicitly captures long-horizon contextual dependencies to enhance precursor sensitivity. Evaluated on six real-world MTS datasets, RED-F achieves substantial improvements in both AP detection accuracy and prediction lead time, demonstrating its effectiveness and generalizability for proactive anomaly forecasting.

Technology Category

• Machine Learning: Time-Series/Data Streams
• Data Mining & Knowledge Management: Anomaly/Outlier Detection
• Planning, Routing, and Scheduling: Activity and Plan Recognition

Application Category

• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• Web Mining and Content Analysis: Large pretrained models with web data
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

The proactive prediction of anomalies (AP) in mul- tivariate time series (MTS) is a critical challenge to ensure system dependability. The difficulty lies in identifying subtle anomaly precursors concealed within normal signals. However, existing unsupervised methods, trained exclusively on normal data, demonstrate a fundamental propensity to reconstruct normal patterns. Consequently, when confronted with weak precursors, their predictions are dominated by the normal pattern, submerging the very signal required for prediction. To contend with the limitation, we propose RED-F, a Reconstruction- Elimination based Dual-stream Contrastive Forecasting frame- work, comprising the Reconstruction-Elimination Model (REM) and the Dual-stream Contrastive Forecasting Model (DFM). The REM utilizes a hybrid time-frequency mechanism to mitigate the precursor, generating a purified, normal-pattern baseline. The DFM then receives this purified baseline and the original sequence which retains the precursor as parallel inputs. At the core of our framework, RED-F employs a contrastive forecast that transforms the difficult task of absolute signal detection into a simpler, more robust task of relative trajectory comparison by computing the divergence between these two predictive streams. This contrastive mechanism serves to amplify the faint precursor signal. Furthermore, the DFM is trained with a novel Multi-Series Prediction (MSP) objective, which leverages distant future con- text to enhance its predictive sensitivity. Extensive experiments on six real-world datasets demonstrate the superior capability of RED-F in anomaly prediction tasks.

Problem

• Identifying subtle anomaly precursors hidden in normal time series signals
• Overcoming unsupervised methods' bias toward reconstructing normal patterns
• Amplifying weak precursor signals for robust anomaly prediction

Innovation

• Reconstruction-Elimination Model purifies normal-pattern baseline
• Dual-stream Contrastive Forecasting compares trajectory divergences
• Multi-Series Prediction leverages distant future context