{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/deep-learning-from-crowds","title":"Deep learning from crowds","arxiv_id":"1709.01779","date":"2017-09-06","proceeding":null,"authors":["Filipe Rodrigues","Francisco Pereira"],"abstract":"Over the last few years, deep learning has revolutionized the field of\nmachine learning by dramatically improving the state-of-the-art in various\ndomains. However, as the size of supervised artificial neural networks grows,\ntypically so does the need for larger labeled datasets. Recently, crowdsourcing\nhas established itself as an efficient and cost-effective solution for labeling\nlarge sets of data in a scalable manner, but it often requires aggregating\nlabels from multiple noisy contributors with different levels of expertise. In\nthis paper, we address the problem of learning deep neural networks from\ncrowds. We begin by describing an EM algorithm for jointly learning the\nparameters of the network and the reliabilities of the annotators. Then, a\nnovel general-purpose crowd layer is proposed, which allows us to train deep\nneural networks end-to-end, directly from the noisy labels of multiple\nannotators, using only backpropagation. We empirically show that the proposed\napproach is able to internally capture the reliability and biases of different\nannotators and achieve new state-of-the-art results for various crowdsourced\ndatasets across different settings, namely classification, regression and\nsequence labeling.","url_abs":"http://arxiv.org/abs/1709.01779v2","url_pdf":"http://arxiv.org/pdf/1709.01779v2.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"deep-learning-from-crowds","repo_url":"https://github.com/NasimISU/OptSLA","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"unanswered"}},{"paper_slug":"deep-learning-from-crowds","repo_url":"https://github.com/friendsqia/friends_qia","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"unanswered"}},{"paper_slug":"deep-learning-from-crowds","repo_url":"https://github.com/ml-postech/robust-deep-learning-from-crowds-with-belief-propagation","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}}],"tasks":[{"task_slug":"deep-learning","task_name":"Deep Learning"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1709.01779","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}