Pay Attention to Evolution: Time Series Forecasting with Deep Graph-Evolution Learning

TL;DR

This paper introduces ReGENN, a novel neural network that combines graph evolution and deep recurrent learning to improve multivariate time-series forecasting by capturing dynamic inter- and intra-temporal relationships, outperforming existing methods.

Contribution

The work presents ReGENN, a new neural architecture that models evolving multivariate relationships in time series, addressing limitations of prior statistical and ensemble approaches.

Findings

ReGENN outperforms dozens of ensemble and statistical methods by up to 64.87%.

Analyzing intermediate weights reveals the importance of simultaneous inter- and intra-temporal relationship modeling.

The approach significantly improves forecasting accuracy by capturing dynamic multivariate dependencies.

Abstract

Time-series forecasting is one of the most active research topics in artificial intelligence. Applications in real-world time series should consider two factors for achieving reliable predictions: modeling dynamic dependencies among multiple variables and adjusting the model's intrinsic hyperparameters. A still open gap in that literature is that statistical and ensemble learning approaches systematically present lower predictive performance than deep learning methods. They generally disregard the data sequence aspect entangled with multivariate data represented in more than one time series. Conversely, this work presents a novel neural network architecture for time-series forecasting that combines the power of graph evolution with deep recurrent learning on distinct data distributions; we named our method Recurrent Graph Evolution Neural Network (ReGENN). The idea is to infer multiple multivariate relationships between co-occurring time-series by assuming that the temporal data depends not only on inner variables and intra-temporal relationships (i.e., observations from itself) but also on outer variables and inter-temporal relationships (i.e., observations from other-selves). An extensive set of experiments was conducted comparing ReGENN with dozens of ensemble methods and classical statistical ones, showing sound improvement of up to 64.87% over the competing algorithms. Furthermore, we present an analysis of the intermediate weights arising from ReGENN, showing that by looking at inter and intra-temporal relationships simultaneously, time-series forecasting is majorly improved if paying attention to how multiple multivariate data synchronously evolve.