Plausible Counterfactuals: Auditing Deep Learning Classifiers with Realistic Adversarial Examples

Simple item page
dc.contributor.authorBarredo-Arrieta, Alejandro
dc.contributor.authorDel Ser, Javier
dc.contributor.institutionTecnalia Research & Innovation
dc.contributor.institutionIA
dc.date.issued2020-07
dc.descriptionPublisher Copyright: © 2020 IEEE.
dc.description.abstractThe last decade has witnessed the proliferation of Deep Learning models in many applications, achieving unrivaled levels of predictive performance. Unfortunately, the black-box nature of Deep Learning models has posed unanswered questions about what they learn from data. Certain application scenarios have highlighted the importance of assessing the bounds under which Deep Learning models operate, a problem addressed by using assorted approaches aimed at audiences from different domains. However, as the focus of the application is placed more on non-expert users, it results mandatory to provide the means for him/her to trust the model, just like a human gets familiar with a system or process: by understanding the hypothetical circumstances under which it fails. This is indeed the angular stone for this research work: to undertake an adversarial analysis of a Deep Learning model. The proposed framework constructs counterfactual examples by ensuring their plausibility, e.g. there is a reasonable probability that a human could generate them without resorting to a computer program. Therefore, this work must be regarded as valuable auditing exercise of the usable bounds a certain model is constrained within, thereby allowing for a much greater understanding of the capabilities and pitfalls of a model used in a real application. To this end, a Generative Adversarial Network (GAN) and multi-objective heuristics are used to furnish a plausible attack to the audited model, efficiently trading between the confusion of this model, the intensity and plausibility of the generated counterfactual. Its utility is showcased within a human face classification task, unveiling the enormous potential of the proposed framework.en
dc.description.sponsorshipThe authors would like to thank the Basque Government for its support through the EMAITEK and ELKARTEK funding programs. Javier Del Ser also receives support from the Consolidated Research Group MATHMODE (IT1294-19) granted by the Department of Education of the Basque Government.
dc.description.statusPeer reviewed
dc.identifier.citationBarredo-Arrieta , A & Del Ser , J 2020 , Plausible Counterfactuals : Auditing Deep Learning Classifiers with Realistic Adversarial Examples . in 2020 International Joint Conference on Neural Networks, IJCNN 2020 - Proceedings . , 9206728 , Proceedings of the International Joint Conference on Neural Networks , Institute of Electrical and Electronics Engineers Inc. , 2020 International Joint Conference on Neural Networks, IJCNN 2020 , Virtual, Glasgow , United Kingdom , 19/07/20 . https://doi.org/10.1109/IJCNN48605.2020.9206728
dc.identifier.citationconference
dc.identifier.doi10.1109/IJCNN48605.2020.9206728
dc.identifier.isbn9781728169262
dc.identifier.urlhttp://www.scopus.com/inward/record.url?scp=85093831442&partnerID=8YFLogxK
dc.language.isoeng
dc.publisherInstitute of Electrical and Electronics Engineers Inc.
dc.relation.ispartof2020 International Joint Conference on Neural Networks, IJCNN 2020 - Proceedings
dc.relation.ispartofseriesProceedings of the International Joint Conference on Neural Networks
dc.relation.projectIDDepartment of Education of the Basque Government
dc.relation.projectIDEusko Jaurlaritza, IT1294-19
dc.rightsinfo:eu-repo/semantics/openAccess
dc.subject.keywordsCounterfactuals
dc.subject.keywordsDeep Learning
dc.subject.keywordsExplainable Artificial Intelligence
dc.subject.keywordsGenerative Adversarial Networks
dc.subject.keywordsMeta-heuristics
dc.subject.keywordsMultiobjective Optimization
dc.subject.keywordsSoftware
dc.subject.keywordsArtificial Intelligence
dc.titlePlausible Counterfactuals: Auditing Deep Learning Classifiers with Realistic Adversarial Examplesen
dc.typeconference output
Files