Show simple item record

dc.contributor.authorBayrak, Betül
dc.contributor.authorBach, Kerstin
dc.date.accessioned2023-12-14T08:26:47Z
dc.date.available2023-12-14T08:26:47Z
dc.date.created2023-11-17T11:38:45Z
dc.date.issued2023
dc.identifier.isbn978-3-031-47993-9
dc.identifier.urihttps://hdl.handle.net/11250/3107501
dc.description.abstractPost-hoc explanation systems offer valuable insights to increase understanding of the predictions made by black-box models. Counterfactual explanations, an instance-based post-hoc explanation method, aim to demonstrate how a model’s prediction can be changed with minimal effort by presenting a hypothetical example. In addition to counterfactual explanation methods, feature attribution techniques such as SHAP (SHapley Additive exPlanations) have also been shown to be effective in providing insights into black-box models. In this paper, we propose PertCF, a perturbation-based counterfactual generation method that benefits from the feature attributions. Our approach combines the strengths of perturbation-based counterfactual generation and feature attribution to generate high-quality, stable, and interpretable counter- factuals. We evaluate PertCF on two open datasets and show that it has promising results over state-of-the-art methods regarding various evalu- ation metrics like stability, proximity, and dissimilarity.en_US
dc.description.abstractPertCF: A Perturbation-Based Counterfactual Generation Approachen_US
dc.language.isoengen_US
dc.publisherSpringeren_US
dc.relation.ispartofArtificial Intelligence XL: 43rd SGAI International Conference on Artificial Intelligence, AI 2023
dc.relation.urihttps://trebuchet.public.springernature.app/get_content/6e85d92d-b190-4934-a30d-10baa975004b?sap-outbound-id=9DE01BA3C8943BB3A05BDE32005A4588D8B5CFCB&utm_source=automation&utm_medium=email&utm_campaig
dc.titlePertCF: A Perturbation-Based Counterfactual Generation Approachen_US
dc.title.alternativePertCF: A Perturbation-Based Counterfactual Generation Approachen_US
dc.typeChapteren_US
dc.description.versionpublishedVersionen_US
dc.rights.holderThis version will not be available due to the publisher's copyright.en_US
dc.source.pagenumber174-187en_US
dc.identifier.doi10.1007/978-3-031-47994-6_13
dc.identifier.cristin2198029
dc.relation.projectNorges forskningsråd: 304843en_US
cristin.ispublishedtrue
cristin.fulltextoriginal
cristin.qualitycode1


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record