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

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. The code is available at https://anonymous.4open.science/status/MultiRC-CCE6. With the advancement of Internet of things (IoT), an increase number of sensors are utilized in industrial facilities to collect data in the form of continuous time series, which realizes the monitoring of system status (Li et al., 2021a). The anomaly detection technology (Li et al., 2021b; Wen et al., 2022; Chen et al., 2021a) has been widely used, which locates system malfunctions by identifying anomalies in historical data (Figure 1a). Effectively detecting anomalies helps pinpoint the sources of faults and prevent the spread of malfunctions. However, anomaly detection can only identify issues after they have occurred, which cannot meet the need for preventive maintenance timely.