GDBN: a Graph Neural Network Approach to Dynamic Bayesian Network
This work addresses causal inference in dynamic systems for applications in science and business analytics, representing an incremental improvement over existing methods.
The paper tackled the problem of identifying causal relations in multi-variate time series by proposing a graph neural network approach to learn sparse dynamic Bayesian networks, demonstrating that it significantly outperforms state-of-the-art methods in dynamic Bayesian network inference.
Identifying causal relations among multi-variate time series is one of the most important elements towards understanding the complex mechanisms underlying the dynamic system. It provides critical tools for forecasting, simulations and interventions in science and business analytics. In this paper, we proposed a graph neural network approach with score-based method aiming at learning a sparse DAG that captures the causal dependencies in a discretized time temporal graph. We demonstrate methods with graph neural network significantly outperformed other state-of-the-art methods with dynamic bayesian networking inference. In addition, from the experiments, the structural causal model can be more accurate than a linear SCM discovered by the methods such as Notears.