Sparsification and Filtering for Spatial-temporal GNN in Multivariate Time-series

TL;DR

This paper introduces an end-to-end spatial-temporal GNN architecture that incorporates a matrix filtering module to improve multivariate time-series prediction by enhancing signal quality and graph sparsification.

Contribution

It presents a novel integration of a filtering module with a GNN for multivariate time-series, explicitly addressing noise and graph sparsification challenges.

Findings

Superior prediction performance over baseline methods

Effective noise reduction through matrix filtering

Robustness across different graph configurations

Abstract

We propose an end-to-end architecture for multivariate time-series prediction that integrates a spatial-temporal graph neural network with a matrix filtering module. This module generates filtered (inverse) correlation graphs from multivariate time series before inputting them into a GNN. In contrast with existing sparsification methods adopted in graph neural network, our model explicitly leverage time-series filtering to overcome the low signal-to-noise ratio typical of complex systems data. We present a set of experiments, where we predict future sales from a synthetic time-series sales dataset. The proposed spatial-temporal graph neural network displays superior performances with respect to baseline approaches, with no graphical information, and with fully connected, disconnected graphs and unfiltered graphs.