A novel extreme adaptive GRU for multivariate time series forecasting

Yifan Zhang,Rui Wu,Sergiu M. Dascalu,Frederick C. Harris
DOI: https://doi.org/10.1038/s41598-024-53460-y
IF: 4.6
2024-02-06
Scientific Reports
Abstract:Multivariate time series forecasting is a critical problem in many real-world scenarios. Recent advances in deep learning have significantly enhanced the ability to tackle such problems. However, a primary challenge in time series forecasting comes from the imbalanced time series data that include extreme events. Despite being a small fraction of the data instances, extreme events can have a negative impact on forecasting as they deviate from the majority. However, many recent time series forecasting methods neglect this issue, leading to suboptimal performance. To address these challenges, we introduce a novel model, the Extreme Event Adaptive Gated Recurrent Unit (eGRU), tailored explicitly for forecasting tasks. The eGRU is designed to effectively learn both normal and extreme event patterns within time series data. Furthermore, we introduce a time series data segmentation technique that divides the input sequence into segments, each comprising multiple time steps. This segmentation empowers the eGRU to capture data patterns at different time step resolutions while simultaneously reducing the overall input length. We conducted comprehensive experiments on four real-world benchmark datasets to evaluate the eGRU's performance. Our results showcase its superiority over vanilla RNNs, LSTMs, GRUs, and other state-of-the-art RNN variants in multivariate time series forecasting. Additionally, we conducted ablation studies to demonstrate the consistently superior performance of eGRU in generating accurate forecasts while incorporating a diverse range of labeling results.
multidisciplinary sciences
What problem does this paper attempt to address?
The paper primarily addresses a key issue in multivariate time series forecasting: how to effectively handle time series data that includes extreme events. Specifically, the paper proposes a new model—the Extreme Event Adaptive Gated Recurrent Unit (eGRU), aimed at improving the prediction capabilities for both normal and extreme events. The main issues mentioned in the paper include: 1. **Imbalanced time series data**: In many real-world scenarios, time series data may contain extreme events, which, although they constitute only a small portion of the data, have a significant impact on the prediction results. 2. **Limitations of existing methods**: Most existing time series forecasting methods are not specifically optimized for extreme events, resulting in poor performance when dealing with such data. 3. **Ignoring the impact of extreme events**: Many studies tend to completely exclude data with extreme events, but this leads to models that are unable to handle extreme situations that may occur in practice. To address these issues, the authors propose the eGRU model, which has the following features: - **Distinguishing between normal and extreme events**: eGRU is designed to model normal and extreme events separately by maintaining two independent hidden states to capture the different patterns of these two types of events. - **Time series segmentation technique**: The input sequence is segmented into multiple segments, each containing multiple time steps. This helps the model capture data patterns at different time resolutions and reduces the overall input length. - **Experimental validation**: Extensive experiments on four real-world benchmark datasets demonstrate the superiority of eGRU in multivariate time series forecasting tasks, especially in handling extreme events. In summary, by proposing the eGRU model, the paper aims to enhance the ability to handle extreme events in time series forecasting tasks, particularly in cases where extreme events have a significant impact on the prediction results.