STTS-EAD: Improving Spatio-Temporal Learning Based Time Series Prediction via
Yuanyuan Liang, Tianhao Zhang, Tingyu Xie
TL;DR
STTS-EAD addresses the challenge of anomalies in multivariate time series forecasting by integrating an Embedded Anomaly Detection (EAD) module into a Spatio-Temporal Learning-based Time Series (STTS) forecasting model. The STTS component constructs temporal and spatial embeddings and learns spatio-temporal features, while the EAD module detects and mitigates anomalies during training, with the two components alternately optimized. Empirical results on a public stock dataset and two real-world coffee-sales datasets show that STTS-EAD outperforms strong baselines, and the gains are amplified when the EAD module is active. This end-to-end anomaly-aware training approach enhances forecasting accuracy and robustness in dynamic MTS settings, enabling better utilization of auxiliary spatio-temporal information and adapting to changing entity counts.
Abstract
Handling anomalies is a critical preprocessing step in multivariate time series prediction. However, existing approaches that separate anomaly preprocessing from model training for multivariate time series prediction encounter significant limitations. Specifically, these methods fail to utilize auxiliary information crucial for identifying latent anomalies associated with spatiotemporal factors during the preprocessing stage. Instead, they rely solely on data distribution for anomaly detection, which can result in the incorrect processing of numerous samples that could otherwise contribute positively to model training. To address this, we propose STTS-EAD, an end-to-end method that seamlessly integrates anomaly detection into the training process of multivariate time series forecasting and aims to improve Spatio-Temporal learning based Time Series prediction via Embedded Anomaly Detection. Our proposed STTS-EAD leverages spatio-temporal information for forecasting and anomaly detection, with the two parts alternately executed and optimized for each other. To the best of our knowledge, STTS-EAD is the first to integrate anomaly detection and forecasting tasks in the training phase for improving the accuracy of multivariate time series forecasting. Extensive experiments on a public stock dataset and two real-world sales datasets from a renowned coffee chain enterprise show that our proposed method can effectively process detected anomalies in the training stage to improve forecasting performance in the inference stage and significantly outperform baselines.