User: Guest  Login
Title:

Conditional Independence in Max-linear Bayesian Networks

Document type:
Zeitungsartikel
Author(s):
Améndola, C., Klüppelberg, C., Lauritzen, S., and Tran, N.
Abstract:
Motivated by extreme value theory, max-linear Bayesian networks have been recently introduced and studied as an alternative to linear structural equation models. However, for max-linear systems the classical independence results for Bayesian networks are far from exhausting valid conditional independence statements. We use tropical linear algebra to derive a compact representation of the conditional distribution given a partial observation, and exploit this to obtain a complete description of al...     »
Keywords:
Bayesian network, conditional independence, directed acyclic graph,d-separation, faithfulness, hitting scenario, Markov properties, max-linear model, tropical geometry
Dewey Decimal Classification:
510 Mathematik
Journal title:
Ann. Appl. Probab.
Year:
2022
Journal volume:
32
Year / month:
2022-02
Quarter:
1. Quartal
Month:
Feb
Journal issue:
1
Pages contribution:
1-45
Language:
en
Fulltext / DOI:
doi:10.1214/21-AAP1670
WWW:
PROJECTeuclid
Publisher:
Institute of Mathematical Statistics
Print-ISSN:
1050-5164
E-ISSN:
2168-8737
Notes:
First available in Project Euclid: 27 February 2022
Status:
Erstveröffentlichung
Submitted:
21.02.2020
Date of publication:
01.02.2022
TUM Institution:
Lehrstuhl für Mathematische Statistik
Format:
Text
 BibTeX