{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,17]],"date-time":"2026-06-17T22:24:06Z","timestamp":1781735046498,"version":"3.54.5"},"reference-count":34,"publisher":"World Scientific Pub Co Pte Ltd","issue":"Supp02","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Int. J. Unc. Fuzz. Knowl. Based Syst."],"published-print":{"date-parts":[[2022,12]]},"abstract":"<jats:p> End users who cannot afford to collect and label big data to train accurate deep learning (DL) models resort to Machine Learning as a Service (MLaaS) providers, who provide paid access to accurate DL models. However, the lack of transparency in how the providers\u2019 models make predictions causes a problem of trust. A way to increase trust (and also to align with ethical regulations) is for predictions to be accompanied by explanations locally and independently generated by the end users (rather than by explanations offered by the model providers). Explanation methods using internal components of DL models (a.k.a. model-specific explanations) are more accurate and effective than those relying solely on the inputs and outputs (a.k.a. model-agnostic explanations). However, end users lack white-box access to the internal components of the providers\u2019 models. To tackle this issue, we propose a novel approach allowing an end user to locally generate model-specific explanations for a DL classification model accessed via a provider\u2019s API. First, we approximate the provider\u2019s model with a local surrogate model. We then use the surrogate model\u2019s components to locally generate model-specific explanations that approximate the explanations obtainable with white-box access to the provider\u2019s DL model. Specifically, we leverage the surrogate model\u2019s gradients to generate adversarial examples that counterfactually explain why an input example is classified into a specific class. Our approach only requires the end user to have unlabeled data of size [Formula: see text] of the provider\u2019s training data and with a similar distribution; given the small size and unlabeled nature of these data, they can be assumed to be already available to the end user or even to be supplied by the provider to build trust in his model. We demonstrate the accuracy and effectiveness of our approach through extensive experiments on two ML tasks: image classification and tabular data classification. The locally generated explanations are consistent with those obtainable with white-box access to the provider\u2019s model, thus giving end users an independent and reliable way to determine if the provider\u2019s model is trustworthy. <\/jats:p>","DOI":"10.1142\/s0218488522400219","type":"journal-article","created":{"date-parts":[[2023,1,27]],"date-time":"2023-01-27T02:47:23Z","timestamp":1674787643000},"page":"255-278","source":"Crossref","is-referenced-by-count":2,"title":["Generating Deep Learning Model-Specific Explanations at the End User\u2019s Side"],"prefix":"10.1142","volume":"30","author":[{"given":"R.","family":"Haffar","sequence":"first","affiliation":[{"name":"Universitat Rovira i Virgili, Department of Computer Engineering and Mathematics, CYBERCAT-Center for Cybersecurity Research of Catalonia, Av. Pa\u00fcos Catalans 26, 43007 Tarragona, Catalonia, Spain"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"N.","family":"Jebreel","sequence":"additional","affiliation":[{"name":"Universitat Rovira i Virgili, Department of Computer Engineering and Mathematics, CYBERCAT-Center for Cybersecurity Research of Catalonia, Av. Pa\u00fcos Catalans 26, 43007 Tarragona, Catalonia, Spain"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"D.","family":"S\u00e1nchez","sequence":"additional","affiliation":[{"name":"Universitat Rovira i Virgili, Department of Computer Engineering and Mathematics, CYBERCAT-Center for Cybersecurity Research of Catalonia, Av. Pa\u00fcos Catalans 26, 43007 Tarragona, Catalonia, Spain"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"J.","family":"Domingo-Ferrer","sequence":"additional","affiliation":[{"name":"Universitat Rovira i Virgili, Department of Computer Engineering and Mathematics, CYBERCAT-Center for Cybersecurity Research of Catalonia, Av. Pa\u00fcos Catalans 26, 43007 Tarragona, Catalonia, Spain"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"219","published-online":{"date-parts":[[2023,1,27]]},"reference":[{"key":"S0218488522400219BIB001","doi-asserted-by":"publisher","DOI":"10.1038\/nature14539"},{"key":"S0218488522400219BIB002","first-page":"1","author":"Dargan S.","year":"2019","journal-title":"Archives of Computational Methods in Engineering"},{"key":"S0218488522400219BIB003","doi-asserted-by":"crossref","first-page":"999","DOI":"10.3390\/app11030999","volume":"11","author":"Jebreel N. M.","year":"2021","journal-title":"Applied Sciences"},{"key":"S0218488522400219BIB004","doi-asserted-by":"crossref","first-page":"896","DOI":"10.1109\/ICMLA.2015.152","volume-title":"2015 IEEE 14th International Conference on Machine Learning and Applications (ICMLA)","author":"Ribeiro M.","year":"2015"},{"key":"S0218488522400219BIB005","doi-asserted-by":"crossref","first-page":"223","DOI":"10.1109\/RE.2019.00032","volume-title":"2019 IEEE 27th International Requirements Engineering Conference (RE)","author":"Chazette L.","year":"2019"},{"key":"S0218488522400219BIB006","doi-asserted-by":"crossref","first-page":"105532","DOI":"10.1016\/j.knosys.2020.105532","volume":"194","author":"Blanco-Justicia A.","year":"2020","journal-title":"Knowledge-Based Systems"},{"key":"S0218488522400219BIB008","doi-asserted-by":"crossref","first-page":"1135","DOI":"10.1145\/2939672.2939778","volume-title":"Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining","author":"Ribeiro M. T.","year":"2016"},{"key":"S0218488522400219BIB009","first-page":"4768","volume-title":"Proceedings of the 31st International Conference on Neural Information Processing Systems","author":"Lundberg S. M.","year":"2017"},{"key":"S0218488522400219BIB010","doi-asserted-by":"publisher","DOI":"10.1371\/journal.pone.0130140"},{"key":"S0218488522400219BIB011","volume-title":"Interpretable Machine Learning","author":"Molnar C.","year":"2020"},{"key":"S0218488522400219BIB013","doi-asserted-by":"crossref","first-page":"239","DOI":"10.1007\/3-540-49430-8_13","volume-title":"Neural Networks: Tricks of the Trade","author":"Simard P. Y.","year":"1998"},{"key":"S0218488522400219BIB015","first-page":"427","volume-title":"Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition","author":"Nguyen A.","year":"2015"},{"key":"S0218488522400219BIB016","volume-title":"Proceedings of the 32th AAAI Conference on Artificial intelligence","volume":"1","author":"Ribeiro M. T.","year":"2018"},{"key":"S0218488522400219BIB019","doi-asserted-by":"crossref","first-page":"839","DOI":"10.1109\/WACV.2018.00097","volume-title":"2018 IEEE Winter Conference on Applications of Computer Vision (WACV)","author":"Chattopadhay A.","year":"2018"},{"key":"S0218488522400219BIB020","first-page":"323","volume-title":"International Conference on Modeling Decisions for Artificial Intelligence","author":"Haffar R.","year":"2021"},{"key":"S0218488522400219BIB021","first-page":"162","volume-title":"Proceedings of the 2017 SIAM International Conference on Data Mining","author":"Lash M. T.","year":"2017"},{"key":"S0218488522400219BIB022","first-page":"100","volume-title":"International Conference on Information Processing and Management of Uncertainty in Knowledge-Based Systems","author":"Laugel T.","year":"2018"},{"key":"S0218488522400219BIB025","volume-title":"International Conference on Learning Representations","author":"Zhang H.","year":"2018"},{"key":"S0218488522400219BIB026","volume-title":"31st Conf. on Neural Information Processing Systems\u2013NIPS 2017","author":"Chen G.","year":"2017"},{"key":"S0218488522400219BIB027","first-page":"325","volume-title":"European Conference on Computer Vision","author":"Wang X.","year":"2020"},{"key":"S0218488522400219BIB028","doi-asserted-by":"crossref","first-page":"1789","DOI":"10.1007\/s11263-021-01453-z","volume":"129","author":"Gou J.","year":"2021","journal-title":"International Journal of Computer Vision"},{"key":"S0218488522400219BIB033","doi-asserted-by":"crossref","first-page":"89","DOI":"10.1007\/978-981-16-7334-4_7","volume-title":"International Conference on Soft Computing in Data Science","author":"Srivastava A.","year":"2021"},{"key":"S0218488522400219BIB034","volume-title":"25th Conference on Neural Information Processing Systems\u2013NIPS 2012","author":"Snoek J.","year":"2012"},{"key":"S0218488522400219BIB035","first-page":"841","volume":"31","author":"Wachter S.","year":"2017","journal-title":"Harvard Journal of Law & Technology"},{"key":"S0218488522400219BIB036","doi-asserted-by":"crossref","first-page":"607","DOI":"10.1145\/3351095.3372850","volume-title":"Proceedings of the 2020 Conference on Fairness, Accountability, and Transparency","author":"Mothilal R. K.","year":"2020"},{"key":"S0218488522400219BIB037","volume-title":"34th Conference in Neural Information Processing Systems\u2013NeurIPS 2020","author":"Jeyakumar J. V.","year":"2020"},{"key":"S0218488522400219BIB038","volume":"34","author":"Slack D.","journal-title":"Advances in Neural Information Processing Systems"},{"key":"S0218488522400219BIB039","first-page":"01","volume-title":"2021 IEEE Symposium Series on Computational Intelligence (SSCI)","author":"Artelt A.","year":"2021"},{"key":"S0218488522400219BIB040","doi-asserted-by":"crossref","first-page":"1135","DOI":"10.1145\/2939672.2939778","volume-title":"Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining","author":"Ribeiro M. T.","year":"2016"},{"key":"S0218488522400219BIB041","first-page":"273","volume-title":"International Conference on Modeling Decisions for Artificial Intelligence","author":"Haffar R.","year":"2020"},{"key":"S0218488522400219BIB043","doi-asserted-by":"publisher","DOI":"10.1007\/3-540-46805-6_19"},{"key":"S0218488522400219BIB045","doi-asserted-by":"crossref","first-page":"829","DOI":"10.1068\/p220829","volume":"22","author":"Brown E.","year":"1993","journal-title":"Perception"},{"key":"S0218488522400219BIB046","doi-asserted-by":"crossref","first-page":"1961","DOI":"10.1016\/S0042-6989(97)00010-2","volume":"37","author":"Fellous J.-M.","year":"1997","journal-title":"Vision Research"},{"key":"S0218488522400219BIB047","doi-asserted-by":"publisher","DOI":"10.1016\/S0031-3203(03)00085-2"}],"container-title":["International Journal of Uncertainty, Fuzziness and Knowledge-Based Systems"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.worldscientific.com\/doi\/pdf\/10.1142\/S0218488522400219","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2023,1,27]],"date-time":"2023-01-27T02:49:05Z","timestamp":1674787745000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.worldscientific.com\/doi\/10.1142\/S0218488522400219"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2022,12]]},"references-count":34,"journal-issue":{"issue":"Supp02","published-print":{"date-parts":[[2022,12]]}},"alternative-id":["10.1142\/S0218488522400219"],"URL":"https:\/\/doi.org\/10.1142\/s0218488522400219","relation":{},"ISSN":["0218-4885","1793-6411"],"issn-type":[{"value":"0218-4885","type":"print"},{"value":"1793-6411","type":"electronic"}],"subject":[],"published":{"date-parts":[[2022,12]]}}}