Pre-training Enhanced Spatial-temporal Graph Neural Network for Multivariate Time Series Forecasting

TL;DR

This paper introduces a pre-training framework called STEP that enhances spatial-temporal graph neural networks for multivariate time series forecasting by capturing long-term temporal patterns, leading to improved prediction accuracy.

Contribution

The paper proposes a novel pre-training approach that enables STGNNs to effectively utilize long-term historical data for better multivariate time series forecasting.

Findings

Significant improvement in forecasting accuracy on three real-world datasets.

Effective learning of long-term temporal patterns through the pre-training model.

Enhanced modeling of dependencies between multiple time series.

Abstract

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.