{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,19]],"date-time":"2026-06-19T20:16:50Z","timestamp":1781900210555,"version":"3.54.5"},"reference-count":43,"publisher":"MDPI AG","issue":"3","license":[{"start":{"date-parts":[[2023,3,22]],"date-time":"2023-03-22T00:00:00Z","timestamp":1679443200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100001665","name":"Artificial Intelligence (AI) Interdisciplinary Institute Artificial and Natural InTelligence Institute (ANITI)","doi-asserted-by":"publisher","award":["ANR-19-PI3A-0004"],"award-info":[{"award-number":["ANR-19-PI3A-0004"]}],"id":[{"id":"10.13039\/501100001665","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100001665","name":"Artificial Intelligence (AI) Interdisciplinary Institute Artificial and Natural InTelligence Institute (ANITI)","doi-asserted-by":"publisher","award":["ANR-21-PRRD-0018"],"award-info":[{"award-number":["ANR-21-PRRD-0018"]}],"id":[{"id":"10.13039\/501100001665","id-type":"DOI","asserted-by":"publisher"}]},{"name":"France Relance","award":["ANR-19-PI3A-0004"],"award-info":[{"award-number":["ANR-19-PI3A-0004"]}]},{"name":"France Relance","award":["ANR-21-PRRD-0018"],"award-info":[{"award-number":["ANR-21-PRRD-0018"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Algorithms"],"abstract":"<jats:p>Automatic recommendation systems based on deep neural networks have become extremely popular during the last decade. Some of these systems can, however, be used in applications that are ranked as High Risk by the European Commission in the AI act\u2014for instance, online job candidate recommendations. When used in the European Union, commercial AI systems in such applications will be required to have proper statistical properties with regard to the potential discrimination they could engender. This motivated our contribution. We present a novel optimal transport strategy to mitigate undesirable algorithmic biases in multi-class neural network classification. Our strategy is model agnostic and can be used on any multi-class classification neural network model. To anticipate the certification of recommendation systems using textual data, we used it on the Bios dataset, for which the learning task consists of predicting the occupation of female and male individuals, based on their LinkedIn biography. The results showed that our approach can reduce undesired algorithmic biases in this context to lower levels than a standard strategy.<\/jats:p>","DOI":"10.3390\/a16030174","type":"journal-article","created":{"date-parts":[[2023,3,22]],"date-time":"2023-03-22T06:00:01Z","timestamp":1679464801000},"page":"174","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":10,"title":["How Optimal Transport Can Tackle Gender Biases in Multi-Class Neural Network Classifiers for Job Recommendations"],"prefix":"10.3390","volume":"16","author":[{"ORCID":"https:\/\/orcid.org\/0009-0006-4920-2406","authenticated-orcid":false,"given":"Fanny","family":"Jourdan","sequence":"first","affiliation":[{"name":"CNRS, Institut de Math\u00e9matiques de Toulouse, Universit\u00e9 de Toulouse, CEDEX 9, F-31062 Toulouse, France"},{"name":"CNRS, Institut de Recherche en Informatique de Toulouse, Universit\u00e9 de Toulouse, CEDEX 9, F-31062 Toulouse, France"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Titon Tshiongo","family":"Kaninku","sequence":"additional","affiliation":[{"name":"CNRS, Institut de Math\u00e9matiques de Toulouse, Universit\u00e9 de Toulouse, CEDEX 9, F-31062 Toulouse, France"},{"name":"AKKODIS Group-Hauts de France, Marcq-en-Baroeul, F-59700 Lille, France"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-7689-8246","authenticated-orcid":false,"given":"Nicholas","family":"Asher","sequence":"additional","affiliation":[{"name":"CNRS, Institut de Recherche en Informatique de Toulouse, Universit\u00e9 de Toulouse, CEDEX 9, F-31062 Toulouse, France"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-1252-2960","authenticated-orcid":false,"given":"Jean-Michel","family":"Loubes","sequence":"additional","affiliation":[{"name":"CNRS, Institut de Math\u00e9matiques de Toulouse, Universit\u00e9 de Toulouse, CEDEX 9, F-31062 Toulouse, France"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-2207-6615","authenticated-orcid":false,"given":"Laurent","family":"Risser","sequence":"additional","affiliation":[{"name":"CNRS, Institut de Math\u00e9matiques de Toulouse, Universit\u00e9 de Toulouse, CEDEX 9, F-31062 Toulouse, France"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1968","published-online":{"date-parts":[[2023,3,22]]},"reference":[{"key":"ref_1","first-page":"1","article-title":"Attention is all you need","volume":"30","author":"Vaswani","year":"2017","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"ref_2","unstructured":"Devlin, J., Chang, M.W., Lee, K., and Toutanova, K. (2018). Bert: Pre-training of deep bidirectional transformers for language understanding. arXiv."},{"key":"ref_3","unstructured":"Radford, A., Narasimhan, K., Salimans, T., and Sutskever, I. (2023, March 14). Improving Language Understanding by Generative Pre-Training. Available online: https:\/\/paperswithcode.com\/paper\/improving-language-understanding-by."},{"key":"ref_4","first-page":"1","article-title":"Sequence to sequence learning with neural networks","volume":"27","author":"Sutskever","year":"2014","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"ref_5","doi-asserted-by":"crossref","first-page":"188","DOI":"10.1080\/00031305.2021.1952897","article-title":"A survey of bias in Machine Learning through the prism of Statistical Parity for the Adult Data Set","volume":"76","author":"Besse","year":"2022","journal-title":"Am. Stat."},{"key":"ref_6","doi-asserted-by":"crossref","unstructured":"De Terwangne, C. (2018). Le R\u00e8glement G\u00e9n\u00e9ral sur la Protection des Donn\u00e9es (RGPD\/GDPR): Analyse Approfondie, Larcier.","DOI":"10.1016\/S1773-035X(18)30207-7"},{"key":"ref_7","doi-asserted-by":"crossref","first-page":"672","DOI":"10.1007\/s10851-022-01090-2","article-title":"Tackling Algorithmic Bias in Neural-Network Classifiers using 2-Wasserstein Regularization","volume":"64","author":"Risser","year":"2022","journal-title":"J. Math. Imaging Vis."},{"key":"ref_8","unstructured":"Liu, Y., Ott, M., Goyal, N., Du, J., Joshi, M., Chen, D., Levy, O., Lewis, M., Zettlemoyer, L., and Stoyanov, V. (2019). Roberta: A robustly optimized bert pretraining approach. arXiv."},{"key":"ref_9","doi-asserted-by":"crossref","unstructured":"De-Arteaga, M., Romanov, A., Wallach, H., Chayes, J., Borgs, C., Chouldechova, A., Geyik, S., Kenthapadi, K., and Kalai, A.T. (2019, January 29\u201331). Bias in bios: A case study of semantic representation bias in a high-stakes setting. Proceedings of the Conference on Fairness, Accountability, and Transparency, Atlanta, GA, USA.","DOI":"10.1145\/3287560.3287572"},{"key":"ref_10","doi-asserted-by":"crossref","unstructured":"Dwork, C., Hardt, M., Pitassi, T., Reingold, O., and Zemel, R. (2012, January 8\u201310). Fairness through awareness. Proceedings of the 3rd innovations in Theoretical Computer Science Conference, Cambridge, MA, USA.","DOI":"10.1145\/2090236.2090255"},{"key":"ref_11","first-page":"1","article-title":"Equality of opportunity in supervised learning","volume":"29","author":"Hardt","year":"2016","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"ref_12","doi-asserted-by":"crossref","unstructured":"Verma, S., and Rubin, J. (2018, January 29). Fairness definitions explained. Proceedings of the 2018 IEEE\/ACM International Workshop on Software Fairness (Fairware), Gothenburg, Sweden.","DOI":"10.1145\/3194770.3194776"},{"key":"ref_13","unstructured":"Kleinberg, J., Mullainathan, S., and Raghavan, M. (2016). Inherent trade-offs in the fair determination of risk scores. arXiv."},{"key":"ref_14","doi-asserted-by":"crossref","first-page":"153","DOI":"10.1089\/big.2016.0047","article-title":"Fair prediction with disparate impact: A study of bias in recidivism prediction instruments","volume":"5","author":"Chouldechova","year":"2017","journal-title":"Big Data"},{"key":"ref_15","unstructured":"Pleiss, G., Raghavan, M., Wu, F., Kleinberg, J., and Weinberger, K.Q. (2017). On fairness and calibration. Adv. Neural Inf. Process. Syst., 30."},{"key":"ref_16","doi-asserted-by":"crossref","unstructured":"Skeem, J., and Lowenkamp, C. (2015). Risk, Race, & Recidivism: Predictive Bias and Disparate Impact (SSRN Scholarly Paper No. ID 2687339), Social Science Research Network.","DOI":"10.2139\/ssrn.2687339"},{"key":"ref_17","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1038\/s41746-020-0288-5","article-title":"Sex and gender differences and biases in artificial intelligence for biomedicine and healthcare","volume":"3","author":"Cirillo","year":"2020","journal-title":"NPJ Digit. Med."},{"key":"ref_18","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1145\/3359301","article-title":"Discrimination through optimization: How Facebook\u2019s Ad delivery can lead to biased outcomes","volume":"3","author":"Ali","year":"2019","journal-title":"Proc. ACM Hum.-Comput. Interact."},{"key":"ref_19","unstructured":"Sapiezynski, P., Ghosh, A., Kaplan, L., Rieke, A., and Mislove, A. (2019). Algorithms that \u201cDon\u2019t See Color\u201d: Comparing Biases in Lookalike and Special Ad Audiences. arXiv."},{"key":"ref_20","doi-asserted-by":"crossref","first-page":"E3635","DOI":"10.1073\/pnas.1720347115","article-title":"Word embeddings quantify 100 years of gender and ethnic stereotypes","volume":"115","author":"Garg","year":"2018","journal-title":"Proc. Natl. Acad. Sci. USA"},{"key":"ref_21","doi-asserted-by":"crossref","unstructured":"Zhao, J., Wang, T., Yatskar, M., Ordonez, V., and Chang, K.W. (2018). Gender bias in coreference resolution: Evaluation and debiasing methods. arXiv.","DOI":"10.18653\/v1\/N18-2003"},{"key":"ref_22","doi-asserted-by":"crossref","first-page":"183","DOI":"10.1126\/science.aal4230","article-title":"Semantics derived automatically from language corpora contain human-like biases","volume":"356","author":"Caliskan","year":"2017","journal-title":"Science"},{"key":"ref_23","doi-asserted-by":"crossref","unstructured":"Gonen, H., and Goldberg, Y. (2019). Lipstick on a pig: Debiasing methods cover up systematic gender biases in word embeddings but do not remove them. arXiv.","DOI":"10.18653\/v1\/N19-1061"},{"key":"ref_24","doi-asserted-by":"crossref","unstructured":"Sikdar, S., Lemmerich, F., and Strohmaier, M. (2022, January 21\u201324). GetFair: Generalized Fairness Tuning of Classification Models. Proceedings of the 2022 ACM Conference on Fairness, Accountability, and Transparency, Seoul, Republic of Korea.","DOI":"10.1145\/3531146.3533094"},{"key":"ref_25","unstructured":"Denis, C., Elie, R., Hebiri, M., and Hu, F. (2021). Fairness guarantee in multi-class classification. arXiv."},{"key":"ref_26","doi-asserted-by":"crossref","unstructured":"Zhang, B.H., Lemoine, B., and Mitchell, M. (2018, January 2\u20133). Mitigating unwanted biases with adversarial learning. Proceedings of the 2018 AAAI\/ACM Conference on AI, Ethics, and Society, New Orleans, LA, USA.","DOI":"10.1145\/3278721.3278779"},{"key":"ref_27","unstructured":"Madras, D., Creager, E., Pitassi, T., and Zemel, R. (2018, January 1\u20133). Learning adversarially fair and transferable representations. Proceedings of the International Conference on Machine Learning. PMLR, New Orleans, LA, USA."},{"key":"ref_28","doi-asserted-by":"crossref","unstructured":"Kamishima, T., Akaho, S., Asoh, H., and Sakuma, J. (2012, January 24\u201328). Fairness-aware classifier with prejudice remover regularizer. Proceedings of the Machine Learning and Knowledge Discovery in Databases: European Conference, ECML PKDD 2012, Bristol, UK.","DOI":"10.1007\/978-3-642-33486-3_3"},{"key":"ref_29","unstructured":"Manisha, P., and Gujar, S. (2018). Fnnc: Achieving fairness through neural networks. arXiv."},{"key":"ref_30","doi-asserted-by":"crossref","unstructured":"Zafar, M.B., Valera, I., Gomez Rodriguez, M., and Gummadi, K.P. (2017, January 3\u20137). Fairness beyond disparate treatment & disparate impact: Learning classification without disparate mistreatment. Proceedings of the 26th International Conference on World Wide Web, Perth, Australia.","DOI":"10.1145\/3038912.3052660"},{"key":"ref_31","unstructured":"Zafar, M.B., Valera, I., Rogriguez, M.G., and Gummadi, K.P. (2017, January 20\u201322). Fairness constraints: Mechanisms for fair classification. Proceedings of the Artificial Intelligence and Statistics. PMLR, Ft. Lauderdale, FL, USA."},{"key":"ref_32","doi-asserted-by":"crossref","unstructured":"Zhao, J., Wang, T., Yatskar, M., Ordonez, V., and Chang, K.W. (2017). Men also like shopping: Reducing gender bias amplification using corpus-level constraints. arXiv.","DOI":"10.18653\/v1\/D17-1323"},{"key":"ref_33","doi-asserted-by":"crossref","first-page":"223","DOI":"10.1137\/16M1080173","article-title":"Optimization Methods for Large-Scale Machine Learning","volume":"60","author":"Bottou","year":"2018","journal-title":"SIAM Rev."},{"key":"ref_34","doi-asserted-by":"crossref","first-page":"541","DOI":"10.1162\/neco.1989.1.4.541","article-title":"Backpropagation Applied to Handwritten Zip Code Recognition","volume":"1","author":"LeCun","year":"1989","journal-title":"Neural Comput."},{"key":"ref_35","first-page":"2257","article-title":"Faster Wasserstein Distance Estimation with the Sinkhorn Divergence","volume":"33","author":"Chizat","year":"2020","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"ref_36","first-page":"1","article-title":"POT: Python Optimal Transport","volume":"22","author":"Flamary","year":"2021","journal-title":"J. Mach. Learn. Res."},{"key":"ref_37","unstructured":"Feydy, J., S\u00e9journ\u00e9, T., Vialard, F.X., Amari, S.i., Trouve, A., and Peyr\u00e9, G. (2019, January 16\u201318). Interpolating between Optimal Transport and MMD using Sinkhorn Divergences. Proceedings of the the 22nd International Conference on Artificial Intelligence and Statistics, Naha, Japan."},{"key":"ref_38","first-page":"385","article-title":"Idiot\u2019s Bayes\u2014Not So Stupid After All?","volume":"69","author":"Hand","year":"2001","journal-title":"Int. Stat. Rev."},{"key":"ref_39","unstructured":"Rish, I. (2001, January 4\u201310). An empirical study of the naive Bayes classifier. Proceedings of the IJCAI 2001 Workshop on Empirical Methods in Artificial Intelligence, Washington, DC, USA."},{"key":"ref_40","doi-asserted-by":"crossref","unstructured":"Zhu, Y., Kiros, R., Zemel, R., Salakhutdinov, R., Urtasun, R., Torralba, A., and Fidler, S. (2015). Aligning Books and Movies: Towards Story-like Visual Explanations by Watching Movies and Reading Books. arXiv.","DOI":"10.1109\/ICCV.2015.11"},{"key":"ref_41","doi-asserted-by":"crossref","unstructured":"Mackenzie, J., Benham, R., Petri, M., Trippas, J.R., Culpepper, J.S., and Moffat, A. (2020, January 19\u201323). CC-News-En: A Large English News Corpus. Proceedings of the CIKM \u201920, 29th ACM International Conference on Information & Knowledge Management, online.","DOI":"10.1145\/3340531.3412762"},{"key":"ref_42","first-page":"9","article-title":"Language models are unsupervised multitask learners","volume":"1","author":"Radford","year":"2019","journal-title":"OpenAI Blog"},{"key":"ref_43","unstructured":"Trinh, T.H., and Le, Q.V. (2018). A simple method for commonsense reasoning. arXiv."}],"container-title":["Algorithms"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/1999-4893\/16\/3\/174\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,10]],"date-time":"2025-10-10T19:00:15Z","timestamp":1760122815000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/1999-4893\/16\/3\/174"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2023,3,22]]},"references-count":43,"journal-issue":{"issue":"3","published-online":{"date-parts":[[2023,3]]}},"alternative-id":["a16030174"],"URL":"https:\/\/doi.org\/10.3390\/a16030174","relation":{},"ISSN":["1999-4893"],"issn-type":[{"value":"1999-4893","type":"electronic"}],"subject":[],"published":{"date-parts":[[2023,3,22]]}}}