Kolmogorov-Arnold Networks for Time Series Granger Causality Inference

TL;DR

This paper introduces KANGCI, a novel neural network architecture that improves Granger causality inference from complex, high-dimensional time series by combining a new model with a time-reversal based causal selection method.

Contribution

The paper presents KANGCI, extending Kolmogorov-Arnold Networks for causal inference, and introduces a time-reversal based algorithm for better causal relationship detection.

Findings

KANGCI achieves competitive accuracy on diverse datasets.

The method effectively handles nonlinear and high-dimensional data.

Time-reversal approach improves causal inference robustness.

Abstract

We propose the Granger causality inference Kolmogorov-Arnold Networks (KANGCI), a novel architecture that extends the recently proposed Kolmogorov-Arnold Networks (KAN) to the domain of causal inference. By extracting base weights from KAN layers and incorporating the sparsity-inducing penalty and ridge regularization, KANGCI effectively infers the Granger causality from time series. Additionally, we propose an algorithm based on time-reversed Granger causality that automatically selects causal relationships with better inference performance from the original or time-reversed time series or integrates the results to mitigate spurious connectivities. Comprehensive experiments conducted on Lorenz-96, Gene regulatory networks, fMRI BOLD signals, VAR, and real-world EEG datasets demonstrate that the proposed model achieves competitive performance to state-of-the-art methods in inferring Granger causality from nonlinear, high-dimensional, and limited-sample time series.