MultiRC: Joint Learning for Time Series Anomaly Prediction and Detection with Multi-scale Reconstructive Contrast

2024/10/21 by Shiyan Hu, Hu, Shiyan, Kai Zhao +11 · 10 citations
Computer Science · Engineering · #Anomaly (physics) #Anomaly Detection Techniques and Applications #Anomaly detection #Artificial intelligence #Cartography #Computer science #Contrast (vision) #Engineering #FOS: Computer and information sciences #Geography #Geology #Joint (building) #Machine Learning (cs.LG) #Machine learning #Network Security and Intrusion Detection #Paleontology #Pattern recognition (psychology) #Physics #Scale (ratio) #Series (stratigraphy) #Time Series Analysis and Forecasting

paper · pdf · doi:10.48550/arxiv.2410.15997

published in arXiv (Cornell University) (Cornell University)

openalex publication_date 2024/10/21 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

Many methods have been proposed for unsupervised time series anomaly detection. Despite some progress, research on predicting future anomalies is still relatively scarce. Predicting anomalies is particularly challenging due to the diverse reaction time and the lack of labeled data. To address these challenges, we propose MultiRC to integrate reconstructive and contrastive learning for joint learning of anomaly prediction and detection, with multi-scale structure and adaptive dominant period mask to deal with the diverse reaction time. MultiRC also generates negative samples to provide essential training momentum for the anomaly prediction tasks and prevent model degradation. We evaluate seven benchmark datasets from different fields. For both anomaly prediction and detection tasks, MultiRC outperforms existing state-of-the-art methods.

Cited by

• RED-F: Reconstruction-Elimination based Dual-stream Contrastive Forecasting for Multivariate Time Series Anomaly Prediction
• ARROW: An Adaptive Rollout and Routing Method for Global Weather Forecasting
• Robust Concept Erasure in Diffusion Models: A Theoretical Perspective on Security and Robustness
• Artificial Intelligence-Based Multiscale Temporal Modeling for Anomaly Detection in Cloud Services
• FADE: Adversarial Concept Erasure in Flow Models
• OFFSET: Segmentation-based Focus Shift Revision for Composed Image Retrieval
• TAB: Unified Benchmarking of Time Series Anomaly Detection Methods
• K²VAE: A Koopman-Kalman Enhanced Variational AutoEncoder for Probabilistic Time Series Forecasting
• Segregation and Context Aggregation Network for Real-time Cloud Segmentation
• AimTS: Augmented Series and Image Contrastive Learning for Time Series Classification

Related

• Unsupervised Time Series Anomaly Prediction with Importance-based Generative Contrastive Learning
• Recurrent Auto-Encoder With Multi-Resolution Ensemble and Predictive Coding for Multivariate Time-Series Anomaly Detection
• A New Perspective on Time Series Anomaly Detection: Faster Patch-based Broad Learning System
• Harnessing Contrastive Learning and Neural Transformation for Time Series Anomaly Detection
• Forecast-based Multi-aspect Framework for Multivariate Time-series Anomaly Detection