English

Interaction Information for Causal Inference: The Case of Directed Triangle

Artificial Intelligence 2017-02-01 v1

Abstract

Interaction information is one of the multivariate generalizations of mutual information, which expresses the amount information shared among a set of variables, beyond the information, which is shared in any proper subset of those variables. Unlike (conditional) mutual information, which is always non-negative, interaction information can be negative. We utilize this property to find the direction of causal influences among variables in a triangle topology under some mild assumptions.

Keywords

Cite

@article{arxiv.1701.08868,
  title  = {Interaction Information for Causal Inference: The Case of Directed Triangle},
  author = {AmirEmad Ghassami and Negar Kiyavash},
  journal= {arXiv preprint arXiv:1701.08868},
  year   = {2017}
}
R2 v1 2026-06-22T18:04:44.602Z