Relevance for robust Bayesian network MAP-explanations

Silja Renooij

Relevance for robust Bayesian network MAP-explanations

Silja Renooij

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

Abstract

In the context of explainable AI, the concept of MAP-independence was recently introduced as a means for conveying the (ir)relevance of intermediate nodes for MAP computations in Bayesian networks. In this paper, we further study the concept of MAP-independence, discuss methods for finding sets of relevant nodes, and suggest ways to use these in providing users with an explanation concerning the robustness of the MAP result.

Original language	English
Title of host publication	Proceedings of The 11th International Conference on Probabilistic Graphical Models, PMLR
Editors	Antonio Salmeron, Rafael Rumi
Publisher	MLResearchPress
Pages	13-24
Publication status	Published - Oct 2022
Event	International Conference on Probabilistic Graphical Models - Almeria, Spain Duration: 5 Oct 2022 → 7 Oct 2022 Conference number: 11

Publication series

Name	Proceedings of Machine Learning Research
Volume	186

Conference

Conference	International Conference on Probabilistic Graphical Models
Abbreviated title	PGM
Country/Territory	Spain
City	Almeria
Period	5/10/22 → 7/10/22

Access to Document

renooij22aFinal published version, 464 KBLicence: Taverne

https://proceedings.mlr.press/v186/renooij22a.html

Cite this

@inproceedings{c692c0c5f89b494dbe516c8df2c063a4,

title = "Relevance for robust Bayesian network MAP-explanations",

abstract = "In the context of explainable AI, the concept of MAP-independence was recently introduced as a means for conveying the (ir)relevance of intermediate nodes for MAP computations in Bayesian networks. In this paper, we further study the concept of MAP-independence, discuss methods for finding sets of relevant nodes, and suggest ways to use these in providing users with an explanation concerning the robustness of the MAP result.",

author = "Silja Renooij",

year = "2022",

month = oct,

language = "English",

series = "Proceedings of Machine Learning Research",

publisher = "MLResearchPress",

pages = "13--24",

editor = "Antonio Salmeron and Rafael Rumi",

booktitle = "Proceedings of The 11th International Conference on Probabilistic Graphical Models, PMLR",

note = "International Conference on Probabilistic Graphical Models, PGM ; Conference date: 05-10-2022 Through 07-10-2022",

}

Renooij, S 2022, Relevance for robust Bayesian network MAP-explanations. in A Salmeron & R Rumi (eds), Proceedings of The 11th International Conference on Probabilistic Graphical Models, PMLR. Proceedings of Machine Learning Research, vol. 186, MLResearchPress, pp. 13-24, International Conference on Probabilistic Graphical Models, Almeria, Spain, 5/10/22. <https://proceedings.mlr.press/v186/renooij22a.html>

Relevance for robust Bayesian network MAP-explanations. / Renooij, Silja.
Proceedings of The 11th International Conference on Probabilistic Graphical Models, PMLR. ed. / Antonio Salmeron; Rafael Rumi. MLResearchPress, 2022. p. 13-24 (Proceedings of Machine Learning Research; Vol. 186).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

TY - GEN

T1 - Relevance for robust Bayesian network MAP-explanations

AU - Renooij, Silja

N1 - Conference code: 11

PY - 2022/10

Y1 - 2022/10

N2 - In the context of explainable AI, the concept of MAP-independence was recently introduced as a means for conveying the (ir)relevance of intermediate nodes for MAP computations in Bayesian networks. In this paper, we further study the concept of MAP-independence, discuss methods for finding sets of relevant nodes, and suggest ways to use these in providing users with an explanation concerning the robustness of the MAP result.

AB - In the context of explainable AI, the concept of MAP-independence was recently introduced as a means for conveying the (ir)relevance of intermediate nodes for MAP computations in Bayesian networks. In this paper, we further study the concept of MAP-independence, discuss methods for finding sets of relevant nodes, and suggest ways to use these in providing users with an explanation concerning the robustness of the MAP result.

M3 - Conference contribution

T3 - Proceedings of Machine Learning Research

SP - 13

EP - 24

BT - Proceedings of The 11th International Conference on Probabilistic Graphical Models, PMLR

A2 - Salmeron, Antonio

A2 - Rumi, Rafael

PB - MLResearchPress

T2 - International Conference on Probabilistic Graphical Models

Y2 - 5 October 2022 through 7 October 2022

ER -

Relevance for robust Bayesian network MAP-explanations

Abstract

Publication series

Conference

Access to Document

Fingerprint

Cite this