Logo do repositório
 
Publicação

Explainability meets uncertainty quantification

2023-12Artigo científico Acesso aberto
Ver registo simples
dc.contributor.authorFolgado, Duarte
dc.contributor.authorBarandas, Marília
dc.contributor.authorFamiglini, Lorenzo
dc.contributor.authorSantos, Ricardo
dc.contributor.authorCabitza, Federico
dc.contributor.authorGamboa, Hugo
dc.contributor.institutionLIBPhys-UNL
dc.contributor.pblElsevier BV
dc.date.accessioned2024-02-24T00:11:55Z
dc.date.available2024-02-24T00:11:55Z
dc.date.issued2023-12
dc.descriptionFunding Information: This work was supported by European funds through the Recovery and Resilience Plan, project “Center for Responsible AI” , project number C645008882-00000055 . Publisher Copyright: © 2023 The Author(s)
dc.description.abstractFeature importance evaluation is one of the prevalent approaches to interpreting Machine Learning (ML) models. A drawback of using these methods for high-dimensional datasets is that they often lead to high-dimensional explanation output that hinders human analysis. This is especially true for explaining multimodal ML models, where the problem's complexity is further exacerbated by the inclusion of multiple data modalities and an increase in the overall number of features. This work proposes a novel approach to lower the complexity of feature-based explanations. The proposed approach is based on uncertainty quantification techniques, allowing for a principled way of reducing the number of modalities required to explain the model's predictions. We evaluated our method in three multimodal datasets comprising physiological time series. Results show that the proposed method can reduce the complexity of the explanations while maintaining a high level of accuracy in the predictions. This study illustrates an innovative example of the intersection between the disciplines of uncertainty quantification and explainable artificial intelligence.en
dc.description.versionpublishersversion
dc.description.versionpublished
dc.format.extent17
dc.format.extent1573680
dc.identifier.doi10.1016/j.inffus.2023.101955
dc.identifier.issn1566-2535
dc.identifier.otherPURE: 83893364
dc.identifier.otherPURE UUID: eacdaeab-b685-46cf-babc-73734d6b6135
dc.identifier.otherScopus: 85169791871
dc.identifier.otherWOS: 001053995800001
dc.identifier.otherORCID: /0000-0002-4022-7424/work/153922997
dc.identifier.urihttp://hdl.handle.net/10362/164087
dc.identifier.urlhttps://www.scopus.com/pages/publications/85169791871
dc.language.isoeng
dc.peerreviewedyes
dc.subjectComplexity
dc.subjectExplainable AI
dc.subjectFeature-based explanations
dc.subjectMultimodal
dc.subjectSHAP
dc.subjectUncertainty quantification
dc.subjectSoftware
dc.subjectSignal Processing
dc.subjectInformation Systems
dc.subjectHardware and Architecture
dc.titleExplainability meets uncertainty quantificationen
dc.title.subtitleInsights from feature-based model fusion on multimodal time seriesen
dc.typejournal article
degois.publication.titleInformation Fusion
degois.publication.volume100
dspace.entity.typePublication
rcaap.rightsopenAccess

Ficheiros

Principais
A mostrar 1 - 1 de 1
Miniatura
Nome:
Explainability_meets_uncertainty_quantification.pdf
Tamanho:
1.5 MB
Formato:
Adobe Portable Document Format
Ver/Abrir

Coleções

Home collection (FCT)