SAGDFN: A Scalable Adaptive Graph Diffusion Forecasting Network for Multivariate Time Series Forecasting

TL;DR

This paper introduces SAGDFN, a scalable graph neural network designed for large-scale multivariate time series forecasting, effectively capturing complex spatial-temporal dependencies without prior spatial knowledge.

Contribution

The paper proposes SAGDFN, a novel scalable adaptive graph diffusion network that handles large datasets with thousands of nodes, overcoming computational limitations of existing methods.

Findings

Achieves comparable performance on a 207-node dataset.

Outperforms state-of-the-art methods on three datasets with 2000 nodes.

Demonstrates scalability and effectiveness for large-scale time series forecasting.

Abstract

Time series forecasting is essential for our daily activities and precise modeling of the complex correlations and shared patterns among multiple time series is essential for improving forecasting performance. Spatial-Temporal Graph Neural Networks (STGNNs) are widely used in multivariate time series forecasting tasks and have achieved promising performance on multiple real-world datasets for their ability to model the underlying complex spatial and temporal dependencies. However, existing studies have mainly focused on datasets comprising only a few hundred sensors due to the heavy computational cost and memory cost of spatial-temporal GNNs. When applied to larger datasets, these methods fail to capture the underlying complex spatial dependencies and exhibit limited scalability and performance. To this end, we present a Scalable Adaptive Graph Diffusion Forecasting Network (SAGDFN) to capture complex spatial-temporal correlation for large-scale multivariate time series and thereby, leading to exceptional performance in multivariate time series forecasting tasks. The proposed SAGDFN is scalable to datasets of thousands of nodes without the need of prior knowledge of spatial correlation. Extensive experiments demonstrate that SAGDFN achieves comparable performance with state-of-the-art baselines on one real-world dataset of 207 nodes and outperforms all state-of-the-art baselines by a significant margin on three real-world datasets of 2000 nodes.