Towards model-agnostic ensemble explanations

2021
book section
conference proceedings
10
dc.abstract.enExplainable Artificial Intelligence (XAI) methods form a large portfolio of different frameworks and algorithms. Although the main goal of all of explanation methods is to provide an insight into the decision process of AI system, their underlying mechanisms may differ. This may result in very different explanations for the same tasks. In this work, we present an approach that aims at combining several XAI algorithms into one ensemble explanation mechanism via quantitative, automated evaluation framework. We focus on model-agnostic explainers to provide most robustness and we demonstrate our approach on image classification task.pl
dc.affiliationWydział Fizyki, Astronomii i Informatyki Stosowanej : Instytut Informatyki Stosowanejpl
dc.conference21st International Conference on Computational Science (ICCS 2021)pl
dc.conference.cityKraków
dc.conference.countryPoland
dc.conference.datefinish2021-06-18
dc.conference.datestart2021-06-16
dc.conference.indexscopustrue
dc.conference.indexwostrue
dc.conference.shortcutInternational Conference on Computational Science
dc.contributor.authorBobek, Szymon - 428058 pl
dc.contributor.authorBałaga, Pawełpl
dc.contributor.authorNalepa, Grzegorz - 200414 pl
dc.contributor.editorPaszyński, Maciejpl
dc.contributor.editorKranzlmüller, Dieterpl
dc.contributor.editorKrzhizhanovskaya, Valeria V.pl
dc.contributor.editorDongarra, Jack J.pl
dc.contributor.editorSloot, Peter M. A.pl
dc.date.accessioned2021-10-25T10:50:07Z
dc.date.available2021-10-25T10:50:07Z
dc.date.issued2021pl
dc.description.conftypeinternationalpl
dc.description.physical39-51pl
dc.description.publication1pl
dc.description.seriesLecture Notes in Computer Science
dc.description.seriesnumber12745
dc.identifier.doi10.1007/978-3-030-77970-2_4pl
dc.identifier.eisbn978-3-030-77970-2pl
dc.identifier.isbn978-3-030-77969-6pl
dc.identifier.projectROD UJ / Opl
dc.identifier.serieseissn1611-3349
dc.identifier.seriesissn0302-9743
dc.identifier.urihttps://ruj.uj.edu.pl/xmlui/handle/item/281522
dc.languageengpl
dc.language.containerengpl
dc.pubinfoCham : Springer International Publishingpl
dc.publisher.ministerialSpringerpl
dc.rightsDodaję tylko opis bibliograficzny*
dc.rights.licencebez licencji
dc.rights.uri*
dc.subject.enexplainable artificial intelligencepl
dc.subject.enmachine learningpl
dc.subject.enimage processingpl
dc.subtypeConferenceProceedingspl
dc.titleTowards model-agnostic ensemble explanationspl
dc.title.containerComputational Science – ICCS 2021 : 21st International Conference, Krakow, Poland, June 16-18, 2021 : proceedings, part IVpl
dc.typeBookSectionpl
dspace.entity.typePublication
dc.abstract.enpl
Explainable Artificial Intelligence (XAI) methods form a large portfolio of different frameworks and algorithms. Although the main goal of all of explanation methods is to provide an insight into the decision process of AI system, their underlying mechanisms may differ. This may result in very different explanations for the same tasks. In this work, we present an approach that aims at combining several XAI algorithms into one ensemble explanation mechanism via quantitative, automated evaluation framework. We focus on model-agnostic explainers to provide most robustness and we demonstrate our approach on image classification task.
dc.affiliationpl
Wydział Fizyki, Astronomii i Informatyki Stosowanej : Instytut Informatyki Stosowanej
dc.conferencepl
21st International Conference on Computational Science (ICCS 2021)
dc.conference.city
Kraków
dc.conference.country
Poland
dc.conference.datefinish
2021-06-18
dc.conference.datestart
2021-06-16
dc.conference.indexscopus
true
dc.conference.indexwos
true
dc.conference.shortcut
International Conference on Computational Science
dc.contributor.authorpl
Bobek, Szymon - 428058
dc.contributor.authorpl
Bałaga, Paweł
dc.contributor.authorpl
Nalepa, Grzegorz - 200414
dc.contributor.editorpl
Paszyński, Maciej
dc.contributor.editorpl
Kranzlmüller, Dieter
dc.contributor.editorpl
Krzhizhanovskaya, Valeria V.
dc.contributor.editorpl
Dongarra, Jack J.
dc.contributor.editorpl
Sloot, Peter M. A.
dc.date.accessioned
2021-10-25T10:50:07Z
dc.date.available
2021-10-25T10:50:07Z
dc.date.issuedpl
2021
dc.description.conftypepl
international
dc.description.physicalpl
39-51
dc.description.publicationpl
1
dc.description.series
Lecture Notes in Computer Science
dc.description.seriesnumber
12745
dc.identifier.doipl
10.1007/978-3-030-77970-2_4
dc.identifier.eisbnpl
978-3-030-77970-2
dc.identifier.isbnpl
978-3-030-77969-6
dc.identifier.projectpl
ROD UJ / O
dc.identifier.serieseissn
1611-3349
dc.identifier.seriesissn
0302-9743
dc.identifier.uri
https://ruj.uj.edu.pl/xmlui/handle/item/281522
dc.languagepl
eng
dc.language.containerpl
eng
dc.pubinfopl
Cham : Springer International Publishing
dc.publisher.ministerialpl
Springer
dc.rights*
Dodaję tylko opis bibliograficzny
dc.rights.licence
bez licencji
dc.rights.uri*
dc.subject.enpl
explainable artificial intelligence
dc.subject.enpl
machine learning
dc.subject.enpl
image processing
dc.subtypepl
ConferenceProceedings
dc.titlepl
Towards model-agnostic ensemble explanations
dc.title.containerpl
Computational Science – ICCS 2021 : 21st International Conference, Krakow, Poland, June 16-18, 2021 : proceedings, part IV
dc.typepl
BookSection
dspace.entity.type
Publication

* The migration of download and view statistics prior to the date of April 8, 2024 is in progress.

Views
30
Views per month
Views per city
Ashburn
6
Krakow
4
Siemianowice Śląskie
3
Beijing
2
Dublin
2
Skalbmierz
2
Warsaw
2
Wroclaw
2
Bydgoszcz
1
Cergy
1

No access

No Thumbnail Available