Interaction Measures, Partition Lattices and Kernel Tests for High-Order Interactions

arXiv2306.00904MaRDI QIDQ6438901

Author name not available (Why is that?)

Publication date: 1 June 2023

Abstract: Models that rely solely on pairwise relationships often fail to capture the complete statistical structure of the complex multivariate data found in diverse domains, such as socio-economic, ecological, or biomedical systems. Non-trivial dependencies between groups of more than two variables can play a significant role in the analysis and modelling of such systems, yet extracting such high-order interactions from data remains challenging. Here, we introduce a hierarchy of $d$-order ($dgeq2$) interaction measures, increasingly inclusive of possible factorisations of the joint probability distribution, and define non-parametric, kernel-based tests to establish systematically the statistical significance of $d$-order interactions. We also establish mathematical links with lattice theory, which elucidate the derivation of the interaction measures and their composite permutation tests; clarify the connection of simplicial complexes with kernel matrix centring; and provide a means to enhance computational efficiency. We illustrate our results numerically with validations on synthetic data, and through an application to neuroimaging data.

Has companion code repository: https://github.com/barahona-research-group/streitberg-interaction

This page was built for publication: Interaction Measures, Partition Lattices and Kernel Tests for High-Order Interactions

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6438901)