Pre-training Enhanced Spatial-temporal Graph Neural Network for Multivariate Time Series Forecasting
This work addresses the problem of capturing long-term patterns in time series forecasting for applications like traffic or energy management, but it is incremental as it builds on existing STGNN methods.
The paper tackles the limitation of Spatial-Temporal Graph Neural Networks (STGNNs) in multivariate time series forecasting by proposing a pre-training model (STEP) that learns from long-term historical data to enhance short-term predictions, with experiments on three datasets showing significant improvements.
Multivariate Time Series (MTS) forecasting plays a vital role in a wide range of applications. Recently, Spatial-Temporal Graph Neural Networks (STGNNs) have become increasingly popular MTS forecasting methods. STGNNs jointly model the spatial and temporal patterns of MTS through graph neural networks and sequential models, significantly improving the prediction accuracy. But limited by model complexity, most STGNNs only consider short-term historical MTS data, such as data over the past one hour. However, the patterns of time series and the dependencies between them (i.e., the temporal and spatial patterns) need to be analyzed based on long-term historical MTS data. To address this issue, we propose a novel framework, in which STGNN is Enhanced by a scalable time series Pre-training model (STEP). Specifically, we design a pre-training model to efficiently learn temporal patterns from very long-term history time series (e.g., the past two weeks) and generate segment-level representations. These representations provide contextual information for short-term time series input to STGNNs and facilitate modeling dependencies between time series. Experiments on three public real-world datasets demonstrate that our framework is capable of significantly enhancing downstream STGNNs, and our pre-training model aptly captures temporal patterns.