Connecting the Dots: Multivariate Time Series Forecasting with Graph Neural Networks

  title={Connecting the Dots: Multivariate Time Series Forecasting with Graph Neural Networks},
  author={Zonghan Wu and Shirui Pan and Guodong Long and Jing Jiang and Xiaojun Chang and Chengqi Zhang},
  journal={Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery \& Data Mining},
  • Zonghan WuShirui Pan Chengqi Zhang
  • Published 24 May 2020
  • Computer Science
  • Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining
Modeling multivariate time series has long been a subject that has attracted researchers from a diverse range of fields including economics, finance, and traffic. A basic assumption behind multivariate time series forecasting is that its variables depend on one another but, upon looking closely, it is fair to say that existing methods fail to fully exploit latent spatial dependencies between pairs of variables. In recent years, meanwhile, graph neural networks (GNNs) have shown high capability… 

Learning Sparse and Continuous Graph Structures for Multivariate Time Series Forecasting

A novel deep learning model that joins graph learning and forecasting, which leverages the spatial information into convolutional operation and extracts temporal dynamics using the diffusion convolution recurrent network and a brand new method named Smooth Sparse Unit (SSU) to learn sparse and continuous graph adjacency matrix.

METRO: A Generic Graph Neural Network Framework for Multivariate Time Series Forecasting

METRO is proposed, a generic framework with multi-scale temporal graphs neural networks, which models the dynamic and cross-scale variable correlations simultaneously simultaneously and provides a modular interpretation of existing GNN-based time series forecasting works as specific instances under the authors' framework.

Balanced Graph Structure Learning for Multivariate Time Series Forecasting

Balanced Graph Structure Learning for Forecasting (BGSLF) is proposed, a novel deep learning model that joins graph structure learning and forecasting that achieves state-of-the-art performances with minor trainable parameters.

Learning the Evolutionary and Multi-scale Graph Structure for Multivariate Time Series Forecasting

A hierarchical graph structure cooperated with the dilated convolution is provided to capture the scale-specific correlations among time series and a series of adjacency matrices are constructed under a recurrent manner to represent the evolving correlations at each layer.

Dynamic Graph Learning-Neural Network for Multivariate Time Series Modeling

A novel framework, namely staticand dynamic-graph learning-neural network (SDGL), which acquires static and dynamic graph matrices from data to model longand short-term patterns respectively and integrates the learned static graph information as inductive bias to construct dynamic graphs and local spatio-temporal patterns better.

Modeling Complex Spatial Patterns with Temporal Features via Heterogenous Graph Embedding Networks

A novel end-to-end deep learning model, termed Multivariate Time Series Forecasting via Heterogenous Graph Neural Networks (MTHetGNN) is proposed, which achieves state-of-the-art results in MTS forecasting task.

Multivariate Time Series Regression with Graph Neural Networks

This work proposes an architecture capable of processing these long sequences in a multivariate time series regression task, using the benefits of Graph Neural Networks to improve predictions, and is tested on two seismic datasets that contain earthquake waveforms.

Expressing Multivariate Time Series as Graphs with Time Series Attention Transformer

The proposed Time Series Attention Transformer (TSAT) represents both temporal information and inter-dependencies of multivariate time series in terms of edge-enhanced dynamic graphs, and applies the embedded dynamic graphs to times series forecasting problems, including two real-world datasets and two benchmark datasets.

Sparse Graph Learning for Spatiotemporal Time Series

By tailoring the gradient estimators to the graph learning problem, the proposed graph learning framework is able to achieve state-of-the-art performance while controlling the sparsity of the learned graph and the computational scalability.

Multivariate Time Series Forecasting with Latent Graph Inference

A new approach for Multivariate Time Series forecasting that jointly infers and leverages relations among time series and its modularity allows it to be integrated with current univariate methods.



DAGCN: Dual Attention Graph Convolutional Networks

DAGCN automatically learns the importance of neighbors at different hops using a novel attention graph convolution layer, and then employs a second attention component, a self-attention pooling layer, to generalize the graph representation from the various aspects of a matrix graph embedding.

Predict then Propagate: Graph Neural Networks meet Personalized PageRank

This paper uses the relationship between graph convolutional networks (GCN) and PageRank to derive an improved propagation scheme based on personalized PageRank, and constructs a simple model, personalized propagation of neural predictions (PPNP), and its fast approximation, APPNP.

Modeling Long- and Short-Term Temporal Patterns with Deep Neural Networks

A novel deep learning framework, namely Long- and Short-term Time-series network (LSTNet), to address this open challenge of multivariate time series forecasting, using the Convolution Neural Network and the Recurrent Neural Network to extract short-term local dependency patterns among variables and to discover long-term patterns for time series trends.

Attention Based Spatial-Temporal Graph Convolutional Networks for Traffic Flow Forecasting

Experiments on two real-world datasets from the Caltrans Performance Measurement System demonstrate that the proposed ASTGCN model outperforms the state-of-the-art baselines.

Diffusion Convolutional Recurrent Neural Network: Data-Driven Traffic Forecasting

Diffusion Convolutional Recurrent Neural Network (DCRNN), a deep learning framework for traffic forecasting that incorporates both spatial and temporal dependency in the traffic flow and evaluates the framework on two real-world large scale road network traffic datasets and observes consistent improvement.

Structured Sequence Modeling with Graph Convolutional Recurrent Networks

The proposed model combines convolutional neural networks on graphs to identify spatial structures and RNN to find dynamic patterns in data structured by an arbitrary graph.

Multi-Range Attentive Bicomponent Graph Convolutional Network for Traffic Forecasting

The Multi-Range Attentive Bicomponent GCN (MRA-BGCN), a novel deep learning model for traffic forecasting that introduces the multi-range attention mechanism to aggregate information in different neighborhood ranges and automatically learn the importance of different ranges.

GMAN: A Graph Multi-Attention Network for Traffic Prediction

Experimental results on two real-world traffic prediction tasks demonstrate the superiority of GMAN, and in the 1 hour ahead prediction, GMAN outperforms state-of-the-art methods by up to 4% improvement in MAE measure.

Spatio-Temporal Graph Convolutional Networks: A Deep Learning Framework for Traffic Forecasting

This paper proposes a novel deep learning framework, Spatio-Temporal Graph Convolutional Networks (STGCN), to tackle the time series prediction problem in traffic domain, and builds the model with complete convolutional structures, which enable much faster training speed with fewer parameters.

Cluster-GCN: An Efficient Algorithm for Training Deep and Large Graph Convolutional Networks

Cluster-GCN is proposed, a novel GCN algorithm that is suitable for SGD-based training by exploiting the graph clustering structure and allows us to train much deeper GCN without much time and memory overhead, which leads to improved prediction accuracy.