Adaptive Dependency Learning Graph Neural Networks
This addresses a bottleneck for using GNNs in domains like retail or energy where predefined graphs are rare, though it is an incremental improvement over existing methods.
The paper tackles the problem of applying Graph Neural Networks (GNNs) to multivariate forecasting when a predefined dependency graph is unavailable, by proposing a hybrid method that self-learns dependencies from data, resulting in significantly improved performance on real-world benchmark datasets.
Graph Neural Networks (GNN) have recently gained popularity in the forecasting domain due to their ability to model complex spatial and temporal patterns in tasks such as traffic forecasting and region-based demand forecasting. Most of these methods require a predefined graph as input, whereas in real-life multivariate time series problems, a well-predefined dependency graph rarely exists. This requirement makes it harder for GNNs to be utilised widely for multivariate forecasting problems in other domains such as retail or energy. In this paper, we propose a hybrid approach combining neural networks and statistical structure learning models to self-learn the dependencies and construct a dynamically changing dependency graph from multivariate data aiming to enable the use of GNNs for multivariate forecasting even when a well-defined graph does not exist. The statistical structure modeling in conjunction with neural networks provides a well-principled and efficient approach by bringing in causal semantics to determine dependencies among the series. Finally, we demonstrate significantly improved performance using our proposed approach on real-world benchmark datasets without a pre-defined dependency graph.