{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,19]],"date-time":"2026-03-19T07:03:41Z","timestamp":1773903821791,"version":"3.50.1"},"reference-count":52,"publisher":"Springer Science and Business Media LLC","issue":"1","license":[{"start":{"date-parts":[[2024,10,14]],"date-time":"2024-10-14T00:00:00Z","timestamp":1728864000000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2024,10,14]],"date-time":"2024-10-14T00:00:00Z","timestamp":1728864000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"funder":[{"DOI":"10.13039\/100014440","name":"Ministerio de Ciencia, Innovaci\u00f3n y Universidades","doi-asserted-by":"publisher","award":["PID2020-118014RB-I00"],"award-info":[{"award-number":["PID2020-118014RB-I00"]}],"id":[{"id":"10.13039\/100014440","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100014440","name":"Ministerio de Ciencia, Innovaci\u00f3n y Universidades","doi-asserted-by":"publisher","award":["PID2022-136627NB-I00\/AEI\/10.13039\/501100011033 FEDER, UE"],"award-info":[{"award-number":["PID2022-136627NB-I00\/AEI\/10.13039\/501100011033 FEDER, UE"]}],"id":[{"id":"10.13039\/100014440","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100017266","name":"Gobierno de Navarra","doi-asserted-by":"publisher","award":["0011-1411-2020-000079 - Emotional Films"],"award-info":[{"award-number":["0011-1411-2020-000079 - Emotional Films"]}],"id":[{"id":"10.13039\/501100017266","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["Prog Artif Intell"],"published-print":{"date-parts":[[2025,3]]},"abstract":"<jats:title>Abstract<\/jats:title>\n          <jats:p>Machine learning models can inherit biases from their training data, leading to discriminatory or inaccurate predictions. This is particularly concerning with the increasing use of large, unsupervised datasets for training foundational models. Traditionally, demographic biases within these datasets have not been well-understood, limiting our ability to understand how they propagate to the models themselves. To address this issue, this paper investigates the propagation of demographic biases from datasets into machine learning models. We focus on the gender demographic component, analyzing two types of bias: representational and stereotypical. For our analysis, we consider the domain of facial expression recognition (FER), a field known to exhibit biases in most popular datasets. We use Affectnet, one of the largest FER datasets, as our baseline for carefully designing and generating subsets that incorporate varying strengths of both representational and stereotypical bias. Subsequently, we train several models on these biased subsets, evaluating their performance on a common test set to assess the propagation of bias into the models\u2019 predictions. Our results show that representational bias has a weaker impact than expected. Models exhibit a good generalization ability even in the absence of one gender in the training dataset. Conversely, stereotypical bias has a significantly stronger impact, primarily concentrated on the biased class, although it can also influence predictions for unbiased classes. These results highlight the need for a bias analysis that differentiates between types of bias, which is crucial for the development of effective bias mitigation strategies.\n<\/jats:p>","DOI":"10.1007\/s13748-024-00345-w","type":"journal-article","created":{"date-parts":[[2024,10,14]],"date-time":"2024-10-14T17:02:47Z","timestamp":1728925367000},"page":"11-31","update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":7,"title":["Less can be more: representational vs. stereotypical gender bias in facial expression recognition"],"prefix":"10.1007","volume":"14","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-6099-8701","authenticated-orcid":false,"given":"Iris","family":"Dominguez-Catena","sequence":"first","affiliation":[]},{"given":"Daniel","family":"Paternain","sequence":"additional","affiliation":[]},{"given":"Aranzazu","family":"Jurio","sequence":"additional","affiliation":[]},{"given":"Mikel","family":"Galar","sequence":"additional","affiliation":[]}],"member":"297","published-online":{"date-parts":[[2024,10,14]]},"reference":[{"key":"345_CR1","doi-asserted-by":"publisher","first-page":"255","DOI":"10.1126\/science.aaa8415","volume":"349","author":"MI Jordan","year":"2015","unstructured":"Jordan, M.I., Mitchell, T.M.: Machine learning: trends, perspectives, and prospects. Science 349, 255\u2013260 (2015)","journal-title":"Science"},{"key":"345_CR2","doi-asserted-by":"publisher","first-page":"50","DOI":"10.1162\/tacl_a_00447","volume":"10","author":"J Kreutzer","year":"2022","unstructured":"Kreutzer, J., et al.: Quality at a glance: an audit of web-crawled multilingual datasets. Trans. Assoc. Comput. Linguist. 10, 50\u201372 (2022)","journal-title":"Trans. Assoc. Comput. Linguist."},{"issue":"2","key":"345_CR3","doi-asserted-by":"publisher","first-page":"2627","DOI":"10.1109\/TPAMI.2022.3169734","volume":"45","author":"Z Zhu","year":"2022","unstructured":"Zhu, Z., et al.: WebFace260M: a benchmark for million-scale deep face recognition. IEEE Trans. Pattern Anal. Mach. Intell. 45(2), 2627\u20132644 (2022)","journal-title":"IEEE Trans. Pattern Anal. Mach. Intell."},{"key":"345_CR4","doi-asserted-by":"crossref","unstructured":"Birhane, A., Prabhu, V.U.: Large image datasets: A pyrrhic win for computer vision?, 1536\u20131546. IEEE, Waikoloa, HI, USA (2021)","DOI":"10.1109\/WACV48630.2021.00158"},{"key":"345_CR5","unstructured":"Brown, T.B., et\u00a0al.: Language Models are Few-Shot Learners. ArXiv (2020). https:\/\/www.semanticscholar.org\/paper\/Language-Models-are-Few-Shot-Learners-Brown-Mann\/6b85b63579a916f705a8e10a49bd8d849d91b1fc"},{"key":"345_CR6","doi-asserted-by":"crossref","unstructured":"Maaz, M., Rasheed, H., Khan, S., Khan, F.S.: Video-ChatGPT: Towards Detailed Video Understanding via Large Vision and Language Models (2023)","DOI":"10.18653\/v1\/2024.acl-long.679"},{"key":"345_CR7","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1145\/3457607","volume":"54","author":"N Mehrabi","year":"2021","unstructured":"Mehrabi, N., Morstatter, F., Saxena, N., Lerman, K., Galstyan, A.: A survey on bias and fairness in machine learning. ACM Comput. Surv. 54, 1\u201335 (2021)","journal-title":"ACM Comput. Surv."},{"key":"345_CR8","unstructured":"Birhane, A., Prabhu, V.U., Kahembwe, E.: Multimodal datasets: Misogyny, pornography, and malignant stereotypes (2021). arxiv:2110.01963"},{"key":"345_CR9","doi-asserted-by":"crossref","unstructured":"Suresh, H., Guttag, J.: A Framework for Understanding Sources of Harm throughout the Machine Learning Life Cycle, 1\u20139 (ACM, \u2013 NY USA, 2021)","DOI":"10.1145\/3465416.3483305"},{"key":"345_CR10","unstructured":"Buolamwini, J., Gebru, T., Friedler, S.A., Wilson, C.: (eds) Gender shades: Intersectional accuracy disparities in commercial gender classification. (eds Friedler, S.\u00a0A. & Wilson, C.) Proceedings of the 1st Conference on Fairness, Accountability and Transparency, Vol.\u00a081 of Proceedings of Machine Learning Research, 77\u201391 (PMLR, 2018-02-23\/2018-02-24). https:\/\/proceedings.mlr.press\/v81\/buolamwini18a.html"},{"key":"345_CR11","doi-asserted-by":"crossref","unstructured":"Garrido-Mu\u00f1oz, I., Mart\u00ednez-Santiago, F., Montejo-R\u00e1ez, A.: MarIA and BETO are sexist: Evaluating gender bias in large language models for Spanish. Lang Resources & Evaluation (2023)","DOI":"10.21203\/rs.3.rs-2256074\/v1"},{"key":"345_CR12","doi-asserted-by":"publisher","first-page":"144","DOI":"10.1111\/1468-2230.12759","volume":"86","author":"J Adams-Prassl","year":"2023","unstructured":"Adams-Prassl, J., Binns, R., Kelly-Lyth, A.: Directly discriminatory algorithms. Mod. Law Rev. 86, 144\u2013175 (2023)","journal-title":"Mod. Law Rev."},{"key":"345_CR13","doi-asserted-by":"publisher","first-page":"30219","DOI":"10.1007\/s10489-023-05052-y","volume":"53","author":"B Verma","year":"2023","unstructured":"Verma, B., Nidhi: From methods to datasets: a detailed study on facial emotion recognition. Appl. Intell. 53, 30219\u201330249 (2023)","journal-title":"Appl. Intell."},{"key":"345_CR14","doi-asserted-by":"publisher","first-page":"529","DOI":"10.1109\/TMM.2021.3128738","volume":"25","author":"L Mou","year":"2023","unstructured":"Mou, L., et al.: Isotropic self-supervised learning for driver drowsiness detection with attention-based multimodal fusion. IEEE Trans. Multimedia 25, 529\u2013542 (2023)","journal-title":"IEEE Trans. Multimedia"},{"issue":"11","key":"345_CR15","doi-asserted-by":"publisher","first-page":"13710","DOI":"10.1007\/s11227-022-04416-4","volume":"78","author":"R Nimmagadda","year":"2022","unstructured":"Nimmagadda, R., Arora, K., Martin, M.V.: Emotion recognition models for companion robots. J. Supercomput. 78(11), 13710\u201313727 (2022)","journal-title":"J. Supercomput."},{"key":"345_CR16","doi-asserted-by":"publisher","first-page":"530","DOI":"10.1109\/TAFFC.2019.2946774","volume":"13","author":"P Werner","year":"2022","unstructured":"Werner, P., et al.: Automatic recognition methods supporting pain assessment: a survey. IEEE Trans. Affect. Comput. 13, 530\u2013552 (2022)","journal-title":"IEEE Trans. Affect. Comput."},{"issue":"2","key":"345_CR17","doi-asserted-by":"publisher","first-page":"641","DOI":"10.1007\/s12530-023-09506-z","volume":"15","author":"B Bakariya","year":"2023","unstructured":"Bakariya, B., et al.: Facial emotion recognition and music recommendation system using CNN-based deep learning techniques. Evol. Syst. 15(2), 641\u2013658 (2023)","journal-title":"Evol. Syst."},{"key":"345_CR18","doi-asserted-by":"crossref","unstructured":"Xu, T., White, J., Kalkan, S., Gunes, H., Bartoli, A., Fusiello, A.: (eds) Investigating Bias and Fairness in Facial Expression Recognition. (eds Bartoli, A. & Fusiello, A.) Computer Vision \u2013 ECCV 2020 Workshops, 506\u2013523 (Springer International Publishing, Cham, 2020)","DOI":"10.1007\/978-3-030-65414-6_35"},{"key":"345_CR19","unstructured":"Domnich, A., Anbarjafari, G.: Responsible AI: Gender bias assessment in emotion recognition. arXiv:2103.11436 [cs] (2021)"},{"key":"345_CR20","doi-asserted-by":"crossref","unstructured":"Deuschel, J., Finzel, B., Rieger, I.: Uncovering the Bias in Facial Expressions. arXiv:2011.11311 [cs] (2021)","DOI":"10.20378\/irb-50304"},{"key":"345_CR21","doi-asserted-by":"crossref","unstructured":"Jannat, S.R., Canavan, S.: Expression Recognition Across Age, 1\u20135 (2021)","DOI":"10.1109\/FG52635.2021.9667062"},{"key":"345_CR22","unstructured":"Poyiadzi, R., Shen, J., Petridis, S., Wang, Y., Pantic, M.: Domain Generalisation for Apparent Emotional Facial Expression Recognition across Age-Groups. arXiv:2110.09168 [cs] (2021)"},{"key":"345_CR23","doi-asserted-by":"crossref","unstructured":"Kim, E., Bryant, D., Srikanth, D., Howard, A.: in Age Bias in Emotion Detection: An Analysis of Facial Emotion Recognition Performance on Young, Middle-Aged, and Older Adults 638\u2013644 (Association for Computing Machinery, New York, NY, USA, 2021). https:\/\/doi.org\/10.1145\/3461702.3462609","DOI":"10.1145\/3461702.3462609"},{"key":"345_CR24","doi-asserted-by":"crossref","unstructured":"Ahmad, K. et\u00a0al.: Comparing the Performance of Facial Emotion Recognition Systems on Real-Life Videos: Gender, Ethnicity and Age. In: Arai, K. (eds.) Proceedings of the Future Technologies Conference (FTC) 2021, Volume 1, Vol. 358 193\u2013210 (Springer International Publishing, Cham, 2022)","DOI":"10.1007\/978-3-030-89906-6_14"},{"key":"345_CR25","unstructured":"Dominguez-Catena, I., Paternain, D., Galar, M.: Assessing Demographic Bias Transfer from Dataset to Model: A Case Study in Facial Expression Recognition (Vienna, Austria, 2022-07-24\/2022-07-25)"},{"key":"345_CR26","doi-asserted-by":"crossref","unstructured":"Dominguez-Catena, I., Paternain, D., Galar, M.: Metrics for Dataset Demographic Bias: A Case Study on Facial Expression Recognition. IEEE Transactions on Pattern Analysis and Machine Intelligence 1\u201318 (2024)","DOI":"10.1109\/TPAMI.2024.3361979"},{"key":"345_CR27","doi-asserted-by":"publisher","first-page":"18","DOI":"10.1109\/TAFFC.2017.2740923","volume":"10","author":"A Mollahosseini","year":"2019","unstructured":"Mollahosseini, A., Hasani, B., Mahoor, M.H.: AffectNet: a database for facial expression, valence, and arousal computing in the wild. IEEE Trans. Affective Comput. 10, 18\u201331 (2019)","journal-title":"IEEE Trans. Affective Comput."},{"key":"345_CR28","doi-asserted-by":"crossref","unstructured":"He, K., Zhang, X., Ren, S., Sun, J.: Deep Residual Learning for Image Recognition 1512, 03385 (2015)","DOI":"10.1109\/CVPR.2016.90"},{"key":"345_CR29","unstructured":"Dosovitskiy, A., et\u00a0al.: An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale (2021). arXiv: 2010.11929"},{"key":"345_CR30","unstructured":"Pessach, D., Shmueli, E.: Algorithmic Fairness. arXiv:2001.09784 [cs, stat] (2020)"},{"key":"345_CR31","doi-asserted-by":"publisher","first-page":"36","DOI":"10.1037\/amp0000972","volume":"78","author":"RN Landers","year":"2022","unstructured":"Landers, R.N., Behrend, T.S.: Auditing the AI auditors: a framework for evaluating fairness and bias in high stakes AI predictive models. Am. Psychol. 78, 36 (2022)","journal-title":"Am. Psychol."},{"key":"345_CR32","doi-asserted-by":"publisher","DOI":"10.1002\/widm.1356","volume":"10","author":"E Ntoutsi","year":"2020","unstructured":"Ntoutsi, E., et al.: Bias in data-driven artificial intelligence systems\u2013An introductory survey. WIREs Data Mining Knowl. Discov. 10, e1356 (2020)","journal-title":"WIREs Data Mining Knowl. Discov."},{"key":"345_CR33","unstructured":"Spokesperson, E.P., Guillot, J.D.: EU AI Act: First regulation on artificial intelligence (2023). https:\/\/www.semanticscholar.org\/paper\/EU-AI-Act%3A-first-regulation-on-artificial-Spokesperson-Guillot\/80527ed02db8fe7574f676ed2aa573eb1ae252a0"},{"key":"345_CR34","doi-asserted-by":"publisher","first-page":"3","DOI":"10.3390\/sci6010003","volume":"6","author":"E Ferrara","year":"2024","unstructured":"Ferrara, E.: Fairness and bias in artificial intelligence: a brief survey of sources, impacts, and mitigation strategies. Science 6, 3 (2024)","journal-title":"Science"},{"key":"345_CR35","doi-asserted-by":"crossref","unstructured":"Fabbrizzi, S., Papadopoulos, S., Ntoutsi, E., Kompatsiaris, I.: A survey on bias in visual datasets. Comput. Vis. Image Underst. 223, 103552 (2022)","DOI":"10.1016\/j.cviu.2022.103552"},{"issue":"2","key":"345_CR36","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1145\/3631326","volume":"1","author":"M Hort","year":"2023","unstructured":"Hort, M., Chen, Z., Zhang, J.M., Harman, M., Sarro, F.: Bias mitigation for machine learning classifiers: a comprehensive survey. ACM J. Responsib. Comput. 1(2), 1\u201352 (2023)","journal-title":"ACM J. Responsib. Comput."},{"key":"345_CR37","unstructured":"Dulhanty, C., Wong, A.: Auditing ImageNet: Towards a Model-driven Framework for Annotating Demographic Attributes of Large-Scale Image Datasets. arXiv:1905.01347 [cs] (2019)"},{"key":"345_CR38","doi-asserted-by":"crossref","unstructured":"Jost, L.: Entropy and diversity. Oikos 113, 363\u2013375 (2006)","DOI":"10.1111\/j.2006.0030-1299.14714.x"},{"key":"345_CR39","unstructured":"Cram\u00e9r, H.: in Chapter 21. The two-dimensional case No.\u00a09 in Princeton Mathematical Series, 282 (Princeton university press, Princeton, 1991)"},{"issue":"6","key":"345_CR40","doi-asserted-by":"publisher","first-page":"867","DOI":"10.1109\/TAI.2022.3159614","volume":"3","author":"G Assuncao","year":"2022","unstructured":"Assuncao, G., Patrao, B., Castelo-Branco, M., Menezes, P.: An overview of emotion in artificial intelligence. IEEE Trans. Artif. Intell. 3(6), 867\u2013886 (2022)","journal-title":"IEEE Trans. Artif. Intell."},{"key":"345_CR41","doi-asserted-by":"publisher","first-page":"124","DOI":"10.1037\/h0030377","volume":"17","author":"P Ekman","year":"1971","unstructured":"Ekman, P., Friesen, W.V.: Constants across cultures in the face and emotion. J. Pers. Soc. Psychol. 17, 124\u2013129 (1971)","journal-title":"J. Pers. Soc. Psychol."},{"key":"345_CR42","doi-asserted-by":"publisher","first-page":"59","DOI":"10.1016\/j.neunet.2014.09.005","volume":"64","author":"IJ Goodfellow","year":"2015","unstructured":"Goodfellow, I.J., et al.: Challenges in representation learning: a report on three machine learning contests. Neural Netw. 64, 59\u201363 (2015)","journal-title":"Neural Netw."},{"key":"345_CR43","doi-asserted-by":"publisher","first-page":"550","DOI":"10.1007\/s11263-017-1055-1","volume":"126","author":"Z Zhang","year":"2018","unstructured":"Zhang, Z., Luo, P., Loy, C.C., Tang, X.: From facial expression recognition to interpersonal relation prediction. Int. J. Comput. Vis. 126, 550\u2013569 (2018)","journal-title":"Int. J. Comput. Vis."},{"key":"345_CR44","doi-asserted-by":"publisher","first-page":"436","DOI":"10.1038\/nature14539","volume":"521","author":"Y LeCun","year":"2015","unstructured":"LeCun, Y., Bengio, Y., Hinton, G.: Deep learning. Nature 521, 436\u2013444 (2015)","journal-title":"Nature"},{"key":"345_CR45","doi-asserted-by":"publisher","first-page":"530","DOI":"10.1037\/pspa0000091","volume":"113","author":"RT Bjornsdottir","year":"2017","unstructured":"Bjornsdottir, R.T., Rule, N.O.: The visibility of social class from facial cues. J. Pers. Soc. Psychol. 113, 530\u2013546 (2017)","journal-title":"J. Pers. Soc. Psychol."},{"key":"345_CR46","doi-asserted-by":"crossref","unstructured":"Hernandez, J., et\u00a0al.: Guidelines for Assessing and Minimizing Risks of Emotion Recognition Applications, 8 (2021)","DOI":"10.1109\/ACII52823.2021.9597452"},{"key":"345_CR47","doi-asserted-by":"crossref","unstructured":"Dominguez-Catena, I., Paternain, D., Galar, M.D.S.A.P.: Analyzing Bias Through Demographic Comparison of Datasets 2312, 14626 (2023)","DOI":"10.2139\/ssrn.4844799"},{"key":"345_CR48","doi-asserted-by":"crossref","unstructured":"Karkkainen, K., Joo, J.: FairFace: Face Attribute Dataset for Balanced Race, Gender, and Age for Bias Measurement and Mitigation, 1547\u20131557. IEEE, Waikoloa, HI, USA (2021)","DOI":"10.1109\/WACV48630.2021.00159"},{"key":"345_CR49","unstructured":"King, D.E.: Max-Margin Object Detection. arXiv:1502.00046 [cs] (2015)"},{"key":"345_CR50","unstructured":"Smith, L.N.: A disciplined approach to neural network hyper-parameters: Part 1 \u2013 learning rate, batch size, momentum, and weight decay. arXiv:1803.09820 [cs, stat] (2018)"},{"key":"345_CR51","doi-asserted-by":"crossref","unstructured":"Dominguez-Catena, I., Paternain, D., Galar, M.: in Gender Stereotyping Impact in Facial Expression Recognition , Vol. 1752 9\u201322 (Springer Nature Switzerland, Cham, 2023)","DOI":"10.1007\/978-3-031-23618-1_1"},{"key":"345_CR52","doi-asserted-by":"publisher","first-page":"1199","DOI":"10.3758\/BF03193553","volume":"67","author":"AP Atkinson","year":"2005","unstructured":"Atkinson, A.P., Tipples, J., Burt, D.M., Young, A.W.: Asymmetric interference between sex and emotion in face perception. Perception & Psychophysics 67, 1199\u20131213 (2005)","journal-title":"Perception & Psychophysics"}],"container-title":["Progress in Artificial Intelligence"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s13748-024-00345-w.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/article\/10.1007\/s13748-024-00345-w\/fulltext.html","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s13748-024-00345-w.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,2,28]],"date-time":"2025-02-28T10:41:12Z","timestamp":1740739272000},"score":1,"resource":{"primary":{"URL":"https:\/\/link.springer.com\/10.1007\/s13748-024-00345-w"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2024,10,14]]},"references-count":52,"journal-issue":{"issue":"1","published-print":{"date-parts":[[2025,3]]}},"alternative-id":["345"],"URL":"https:\/\/doi.org\/10.1007\/s13748-024-00345-w","relation":{},"ISSN":["2192-6352","2192-6360"],"issn-type":[{"value":"2192-6352","type":"print"},{"value":"2192-6360","type":"electronic"}],"subject":[],"published":{"date-parts":[[2024,10,14]]},"assertion":[{"value":"8 April 2024","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"14 September 2024","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"14 October 2024","order":3,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}}]}}