Quantifying intrinsic causal contributions via structure preserving interventions

arXiv:2007.00714v4 Announce Type: replace-cross
Abstract: We propose a notion of causal influence that describes the `intrinsic' part of the contribution of a node on a target node in a DAG. By recursively writing each node as a function of the upstream noise terms, we separate the intrinsic information added by each node from the one obtained from its ancestors. To interpret the intrinsic information as a {\it causal} contribution, we consider `structure-preserving interventions' that randomize each node in a way …

abstract arxiv causal cs.ai cs.it dag function influence information intrinsic math.it node noise notion part stat.ml terms type via writing