Temporal Attention Evolutional Graph Convolutional Network for Multivariate Time Series Forecasting

TL;DR

This paper introduces TAEGCN, a novel neural network that captures dynamic spatial-temporal dependencies in multivariate time series data using attention mechanisms and evolving graph structures, improving forecasting accuracy.

Contribution

The paper presents TAEGCN, a new model that dynamically learns graph structures and temporal features for multivariate time series forecasting, addressing limitations of fixed graph assumptions.

Findings

Outperforms existing models on METR-LA and PEMS-BAY datasets.

Effectively captures dynamic spatial-temporal dependencies.

Demonstrates superior forecasting accuracy.

Abstract

Multivariate time series forecasting enables the prediction of future states by leveraging historical data, thereby facilitating decision-making processes. Each data node in a multivariate time series encompasses a sequence of multiple dimensions. These nodes exhibit interdependent relationships, forming a graph structure. While existing prediction methods often assume a fixed graph structure, many real-world scenarios involve dynamic graph structures. Moreover, interactions among time series observed at different time scales vary significantly. To enhance prediction accuracy by capturing precise temporal and spatial features, this paper introduces the Temporal Attention Evolutional Graph Convolutional Network (TAEGCN). This novel method not only integrates causal temporal convolution and a multi-head self-attention mechanism to learn temporal features of nodes, but also construct the dynamic graph structure based on these temporal features to keep the consistency of the changing in spatial feature with temporal series. TAEGCN adeptly captures temporal causal relationships and hidden spatial dependencies within the data. Furthermore, TAEGCN incorporates a unified neural network that seamlessly integrates these components to generate final predictions. Experimental results conducted on two public transportation network datasets, METR-LA and PEMS-BAY, demonstrate the superior performance of the proposed model.