In this paper, we tackle structure learning of Directed Acyclic Graphs (DAGs), with the idea of exploiting available prior knowledge of the domain at hand to guide the search of 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 several popular competitors, to study its behaviour in finite samples. Biological validation of the algorithm is presented through the analysis of non-small cell lung cancer data.
Thi Kim Hue Nguyen, Monica Chiogna, Davide Risso, Erika Banzato (In stampa/Attività in corso). Guided structure learning of DAGs for count data. STATISTICAL MODELLING, To be defined, 1-25 [10.1177/1471082X241266738].
Guided structure learning of DAGs for count data
Monica ChiognaCo-primo
;
In corso di stampa
Abstract
In this paper, we tackle structure learning of Directed Acyclic Graphs (DAGs), with the idea of exploiting available prior knowledge of the domain at hand to guide the search of 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 several popular competitors, to study its behaviour in finite samples. Biological validation of the algorithm is presented through the analysis of non-small cell lung cancer data.File | Dimensione | Formato | |
---|---|---|---|
Guided_structure_learning_of_DAGs_for_count_data.pdf
accesso aperto
Descrizione: AAM
Tipo:
Postprint
Licenza:
Licenza per Accesso Aperto. Altra tipologia di licenza compatibile con Open Access
Dimensione
588.89 kB
Formato
Adobe PDF
|
588.89 kB | Adobe PDF | Visualizza/Apri |
nguyen-et-al-2024-guided-structure-learning-of-dags-for-count-data.pdf
accesso aperto
Descrizione: online first
Tipo:
Versione (PDF) editoriale
Licenza:
Licenza per Accesso Aperto. Creative Commons Attribuzione (CCBY)
Dimensione
317.48 kB
Formato
Adobe PDF
|
317.48 kB | Adobe PDF | Visualizza/Apri |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.