Unsupervised Time Series Anomaly Prediction with Importance-based Generative Contrastive Learning

2024-10-22

arXiv.org

AI Summary

This paper addresses unsupervised time-series anomaly *precursor* prediction—i.e., early detection of anomalous precursors without labeled data and with generalization to unseen anomaly types. We propose the first precursor-oriented framework for this task. Our method introduces three key innovations: (1) an anomaly-precursor disentanglement mechanism that explicitly separates normal temporal evolution from precursor dynamics; (2) an importance-weighted memory bank that adaptively stores and retrieves discriminative historical patterns; and (3) a synergistic optimization of generative contrastive learning and theory-guided unsupervised representation learning. Evaluated on seven benchmark datasets, our approach significantly outperforms state-of-the-art methods, achieving an average 4.2% improvement in F1-score. Notably, it demonstrates superior generalization to previously unseen anomalies and robust predictive performance under zero-shot anomaly conditions.

Technology Category

• Machine Learning: Unsupervised & Self-Supervised Learning
• Data Mining & Knowledge Management: Anomaly/Outlier Detection
• Search and Optimization: Learning to Search

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• 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

Abstract

Time series anomaly prediction plays an essential role in many real-world scenarios, such as environmental prevention and prompt maintenance of cyber-physical systems. However, existing time series anomaly prediction methods mainly require supervised training with plenty of manually labeled data, which are difficult to obtain in practice. Besides, unseen anomalies can occur during inference, which could differ from the labeled training data and make these models fail to predict such new anomalies. In this paper, we study a novel problem of unsupervised time series anomaly prediction. We provide a theoretical analysis and propose Importance-based Generative Contrastive Learning (IGCL) to address the aforementioned problems. IGCL distinguishes between normal and anomaly precursors, which are generated by our anomaly precursor pattern generation module. To address the efficiency issues caused by the potential complex anomaly precursor combinations, we propose a memory bank with importance-based scores to adaptively store representative anomaly precursors and generate more complicated anomaly precursors. Extensive experiments on seven benchmark datasets show our method outperforms state-of-the-art baselines on unsupervised time series anomaly prediction problems.

Problem

• Unsupervised prediction of time series anomalies without labeled data.
• Handling unseen anomalies differing from labeled training data.
• Improving efficiency in detecting complex anomaly precursor combinations.

Innovation

• Unsupervised anomaly prediction using generative contrastive learning
• Anomaly precursor pattern generation module
• Memory bank with importance-based scores