A0520
Title: Guided structure learning of DAGs for count data
Authors: Thi Kim Hue Nguyen - University of Padova (Italy) [presenting]
Monica Chiogna - University of Bologna (Italy)
Davide Risso - University of Padua (Italy)
Erika Banzato - University of Padova (Italy)
Abstract: Structure learning of Directed Acyclic Graphs (DAGs) is tackled, with the idea of exploiting available prior knowledge of the domain at hand to guide the search for the best structure. In particular, we assume to know the topological ordering of variables in addition to the given data. We study a new algorithm for learning the structure of DAGs, proving its theoretical consistency in the limit of infinite observations. Furthermore, we experimentally compare the proposed algorithm to a number of popular competitors, in order to study its behavior in finite samples
