Jacobian Regularizer-based Neural Granger Causality

TL;DR

This paper introduces JRNGC, a novel neural Granger causality method using Jacobian regularization, enabling efficient, accurate, and comprehensive causality analysis with a single model.

Contribution

The paper proposes a Jacobian regularizer-based approach that models multivariate and full-time Granger causality with lower complexity, improving accuracy over existing methods.

Findings

Achieves competitive performance with state-of-the-art methods

Maintains lower model complexity and high scalability

Effectively models full-time Granger causality

Abstract

With the advancement of neural networks, diverse methods for neural Granger causality have emerged, which demonstrate proficiency in handling complex data, and nonlinear relationships. However, the existing framework of neural Granger causality has several limitations. It requires the construction of separate predictive models for each target variable, and the relationship depends on the sparsity on the weights of the first layer, resulting in challenges in effectively modeling complex relationships between variables as well as unsatisfied estimation accuracy of Granger causality. Moreover, most of them cannot grasp full-time Granger causality. To address these drawbacks, we propose a Jacobian Regularizer-based Neural Granger Causality (JRNGC) approach, a straightforward yet highly effective method for learning multivariate summary Granger causality and full-time Granger causality by constructing a single model for all target variables. Specifically, our method eliminates the sparsity constraints of weights by leveraging an input-output Jacobian matrix regularizer, which can be subsequently represented as the weighted causal matrix in the post-hoc analysis. Extensive experiments show that our proposed approach achieves competitive performance with the state-of-the-art methods for learning summary Granger causality and full-time Granger causality while maintaining lower model complexity and high scalability.