Logo do repositório
 
Publicação

On Modifying the Perception of a Neural Network

dc.contributor.authorDe Sousa Ribeiro, Manuel
dc.contributor.authorLeite, João
dc.contributor.institutionNOVALincs
dc.contributor.pblUniversidada de Los Lagos
dc.date.accessioned2026-03-03T15:57:01Z
dc.date.available2026-03-03T15:57:01Z
dc.date.issued2025-10-09
dc.descriptionPublisher Copyright: © 2025 Copyright for this paper by its authors.
dc.description.abstractArtificial neural networks are typically regarded as black boxes, given how difficult it is for humans to interpret how these models reach their results. One way in which humans attempt to interpret complex systems is by imagining how they behave in hypothetical scenarios. In this work, we propose a method that allows one to modify what an artificial neural network is perceiving regarding specific human-defined concepts of interest, allowing one to test how they behave under such hypothetical scenarios. Through empirical evaluation, we test the proposed method on different models and datasets, assessing its qualities.en
dc.description.versionpublishersversion
dc.description.versionpublished
dc.format.extent15
dc.format.extent2097961
dc.identifier.issn1613-0073
dc.identifier.otherPURE: 155627130
dc.identifier.otherPURE UUID: d2e764c3-ec43-41dd-8d5c-1c589fdec097
dc.identifier.otherScopus: 105019517337
dc.identifier.urihttp://hdl.handle.net/10362/200912
dc.identifier.urlhttps://www.scopus.com/pages/publications/105019517337
dc.language.isoeng
dc.peerreviewedyes
dc.subjectCounterfactual
dc.subjectExplainability
dc.subjectInterpretability
dc.subjectNeural Networks
dc.subjectGeneral Computer Science
dc.titleOn Modifying the Perception of a Neural Networken
dc.typejournal article
degois.publication.titleCEUR Workshop Proceedings
degois.publication.volume4061
dspace.entity.typePublication
rcaap.rightsopenAccess

Ficheiros

Principais
A mostrar 1 - 1 de 1
A carregar...
Miniatura
Nome:
Ribeiro_M._Leite_J._2025._On_modifying_the_perception_of_a_neural_network..pdf
Tamanho:
2 MB
Formato:
Adobe Portable Document Format