{"id":652367,"date":"2025-02-05T22:55:15","date_gmt":"2025-02-05T19:55:15","guid":{"rendered":"https:\/\/en.buradabiliyorum.com\/deepfake-detection-improves-when-using-algorithms-that-are-more-aware-of-demographic-diversity\/"},"modified":"2025-02-05T22:55:15","modified_gmt":"2025-02-05T19:55:15","slug":"deepfake-detection-improves-when-using-algorithms-that-are-more-aware-of-demographic-diversity","status":"publish","type":"post","link":"https:\/\/buradabiliyorum.com\/en\/deepfake-detection-improves-when-using-algorithms-that-are-more-aware-of-demographic-diversity\/","title":{"rendered":"#Deepfake detection improves when using algorithms that are more aware of demographic diversity"},"content":{"rendered":"<div>\n<p>Deepfakes \u2013 essentially putting words in someone else\u2019s mouth in a very believable way \u2013 <a rel=\"nofollow\" target=\"_blank\" href=\"https:\/\/theconversation.com\/text-to-image-ai-powerful-easy-to-use-technology-for-making-art-and-fakes-195517\" target=\"_blank\" rel=\"nofollow noopener\">are becoming more sophisticated<\/a> by the day and increasingly hard to spot. Recent examples of deepfakes include <a rel=\"nofollow\" target=\"_blank\" href=\"https:\/\/www.axios.com\/2024\/02\/03\/taylor-swift-deepfake-ai-image-protection\" target=\"_blank\" rel=\"nofollow noopener\">Taylor Swift nude images<\/a>, an <a rel=\"nofollow\" target=\"_blank\" href=\"https:\/\/www.bloomberg.com\/news\/articles\/2024-01-26\/ai-startup-elevenlabs-bans-account-blamed-for-biden-audio-deepfake\" target=\"_blank\" rel=\"nofollow noopener\">audio recording of President Joe Biden<\/a> telling New Hampshire residents not to vote, and a <a rel=\"nofollow\" target=\"_blank\" href=\"https:\/\/www.npr.org\/2022\/03\/16\/1087062648\/deepfake-video-zelenskyy-experts-war-manipulation-ukraine-russia\" target=\"_blank\" rel=\"nofollow noopener\">video of Ukrainian President Volodymyr Zelenskyy<\/a> calling on his troops to lay down their arms.<\/p>\n<p>Although companies have created detectors to help spot deepfakes, studies have found that <a rel=\"nofollow\" target=\"_blank\" href=\"https:\/\/doi.org\/10.48550\/arXiv.2105.00558\" target=\"_blank\" rel=\"nofollow noopener\">biases in the data<\/a> used to train these tools can lead to certain demographic groups being unfairly targeted.<\/p>\n<figure class=\"align-center \"><img decoding=\"async\" src=\"https:\/\/images.theconversation.com\/files\/584931\/original\/file-20240328-20-vx3sfw.jpeg?ixlib=rb-4.1.0&amp;rect=224%2C52%2C4767%2C3270&amp;q=45&amp;auto=format&amp;w=754&amp;fit=clip\" sizes=\"(min-width: 1466px) 754px, (max-width: 599px) 100vw, (min-width: 600px) 600px, 237px\" srcset=\"https:\/\/images.theconversation.com\/files\/584931\/original\/file-20240328-20-vx3sfw.jpeg?ixlib=rb-4.1.0&amp;q=45&amp;auto=format&amp;w=600&amp;h=400&amp;fit=crop&amp;dpr=1 600w, https:\/\/images.theconversation.com\/files\/584931\/original\/file-20240328-20-vx3sfw.jpeg?ixlib=rb-4.1.0&amp;q=30&amp;auto=format&amp;w=600&amp;h=400&amp;fit=crop&amp;dpr=2 1200w, https:\/\/images.theconversation.com\/files\/584931\/original\/file-20240328-20-vx3sfw.jpeg?ixlib=rb-4.1.0&amp;q=15&amp;auto=format&amp;w=600&amp;h=400&amp;fit=crop&amp;dpr=3 1800w, https:\/\/images.theconversation.com\/files\/584931\/original\/file-20240328-20-vx3sfw.jpeg?ixlib=rb-4.1.0&amp;q=45&amp;auto=format&amp;w=754&amp;h=503&amp;fit=crop&amp;dpr=1 754w, https:\/\/images.theconversation.com\/files\/584931\/original\/file-20240328-20-vx3sfw.jpeg?ixlib=rb-4.1.0&amp;q=30&amp;auto=format&amp;w=754&amp;h=503&amp;fit=crop&amp;dpr=2 1508w, https:\/\/images.theconversation.com\/files\/584931\/original\/file-20240328-20-vx3sfw.jpeg?ixlib=rb-4.1.0&amp;q=15&amp;auto=format&amp;w=754&amp;h=503&amp;fit=crop&amp;dpr=3 2262w\" alt=\"a hand holds a smartphone with text on it in front of a screen with a man in front of a lectern\"\/><figcaption><span class=\"caption\">A deepfake of Ukraine President Volodymyr Zelensky in 2022 purported to show him calling on his troops to lay down their arms.<\/span><br \/><span class=\"attribution\"><a rel=\"nofollow\" target=\"_blank\" class=\"source\" href=\"https:\/\/www.gettyimages.com\/detail\/news-photo\/this-illustration-photo-taken-on-january-30-2023-shows-a-news-photo\/1246676884?adppopup=true\" target=\"_blank\" rel=\"nofollow noopener\">Olivier Douliery\/AFP via Getty Images<\/a><\/span><\/figcaption><\/figure>\n<p>My team and I discovered new methods that improve both the fairness and the accuracy of the algorithms used to detect deepfakes.<\/p>\n<p>To do so, we used a large dataset of facial forgeries that lets researchers like us train our deep-learning <a href=\"https:\/\/buradabiliyorum.com\/en\/category\/download-scripts-themes-apps\/\" data-internallinksmanager029f6b8e52c=\"9\" title=\"Download Scripts &amp; Themes &amp; Apps\" target=\"_blank\" rel=\"noopener\">app<\/a>roaches. We built our work around the state-of-the-art Xception detection algorithm, which is a <a rel=\"nofollow\" target=\"_blank\" href=\"https:\/\/doi.org\/10.48550\/arXiv.1901.08971\" target=\"_blank\" rel=\"nofollow noopener\">widely used foundation<\/a> for deepfake detection systems and can detect deepfakes with an accuracy of 91.5%.<\/p>\n<p><a rel=\"nofollow\" target=\"_blank\" href=\"https:\/\/openaccess.thecvf.com\/content\/WACV2024\/papers\/Ju_Improving_Fairness_in_Deepfake_Detection_WACV_2024_paper.pdf\" target=\"_blank\" rel=\"nofollow noopener\">We created two separate deepfake detection methods<\/a> intended to encourage fairness.<\/p>\n<p>One was focused on making the algorithm more aware of demographic diversity by labeling datasets by gender and race to minimize errors among underrepresented groups.<\/p>\n<p>The other aimed to improve fairness without relying on demographic labels by focusing instead on features not visible to the human eye.<\/p>\n<p>It turns out the first method worked best. It increased accuracy rates from the 91.5% baseline to 94.17%, which was a bigger increase than our second method as well as several others we tested. Moreover, it increased accuracy while enhancing fairness, which was our main focus.<\/p>\n<p>We believe fairness and accuracy are crucial if the public is to accept artificial intelligence <a href=\"https:\/\/buradabiliyorum.com\/en\/category\/technology\/\" data-internallinksmanager029f6b8e52c=\"4\" title=\"Technology\" target=\"_blank\" rel=\"noopener\">technology<\/a>. When large language models like ChatGPT \u201challucinate,\u201d <a rel=\"nofollow\" target=\"_blank\" href=\"https:\/\/theconversation.com\/chatgpt-and-other-generative-ai-could-foster-science-denial-and-misunderstanding-heres-how-you-can-be-on-alert-204897\" target=\"_blank\" rel=\"nofollow noopener\">they can perpetuate erroneous information<\/a>. This affects public trust and safety.<\/p>\n<p>Likewise, deepfake images and videos can undermine the adoption of AI if they cannot be quickly and accurately detected. Improving the fairness of these detection algorithms so that certain demographic groups aren\u2019t disproportionately harmed by them is a key aspect to this.<\/p>\n<p>Our research addresses deepfake detection algorithms\u2019 fairness, rather than just attempting to balance the data. It offers a new approach to algorithm design that considers demographic fairness as a core aspect.<!-- Below is The Conversation's page counter tag. Please DO NOT REMOVE. --><img loading=\"lazy\" decoding=\"async\" style=\"border: none !important; box-shadow: none !important; margin: 0 !important; max-height: 1px !important; max-width: 1px !important; min-height: 1px !important; min-width: 1px !important; opacity: 0 !important; outline: none !important; padding: 0 !important;\" alt=\"The Conversation\" width=\"1\" height=\"1\" class=\"js-lazy\" src=\"https:\/\/counter.theconversation.com\/content\/226061\/count.gif?distributor=republish-lightbox-basic\"\/><!-- End of code. If you don't see any code above, please get new code from the Advanced tab after you click the republish button. The page counter does not collect any personal data. More info: https:\/\/theconversation.com\/republishing-guidelines --><img loading=\"lazy\" decoding=\"async\" style=\"border: none !important; box-shadow: none !important; margin: 0 !important; max-height: 1px !important; max-width: 1px !important; min-height: 1px !important; min-width: 1px !important; opacity: 0 !important; outline: none !important; padding: 0 !important;\" src=\"https:\/\/counter.theconversation.com\/content\/226061\/count.gif?distributor=republish-lightbox-basic\" alt=\"The Conversation\" width=\"1\" height=\"1\" class=\"\" srcset=\"\"\/><\/p>\n<p><em><a rel=\"nofollow\" target=\"_blank\" href=\"https:\/\/theconversation.com\/profiles\/siwei-lyu-528645\" target=\"_blank\" rel=\"nofollow noopener\">Siwei Lyu<\/a>, Professor of Computer <a href=\"https:\/\/buradabiliyorum.com\/en\/category\/sciencee\/\" data-internallinksmanager029f6b8e52c=\"5\" title=\"Science\" target=\"_blank\" rel=\"noopener\">Science<\/a> and Engineering; Director, UB <a href=\"https:\/\/buradabiliyorum.com\/en\/category\/social-mediaa\/\" data-internallinksmanager029f6b8e52c=\"1\" title=\"Social Media\" target=\"_blank\" rel=\"noopener\">Media<\/a> Forensic Lab, <a rel=\"nofollow\" target=\"_blank\" href=\"https:\/\/theconversation.com\/institutions\/university-at-buffalo-925\" target=\"_blank\" rel=\"nofollow noopener\">University at Buffalo<\/a> and <a rel=\"nofollow\" target=\"_blank\" href=\"https:\/\/theconversation.com\/profiles\/yan-ju-1523096\" target=\"_blank\" rel=\"nofollow noopener\">Yan Ju<\/a>, Ph.D. Candidate in Computer Science and Engineering, <a rel=\"nofollow\" target=\"_blank\" href=\"https:\/\/theconversation.com\/institutions\/university-at-buffalo-925\" target=\"_blank\" rel=\"nofollow noopener\">University at Buffalo<\/a><\/em><\/p>\n<p><em>This article is republished from <a rel=\"nofollow\" target=\"_blank\" href=\"https:\/\/theconversation.com\" target=\"_blank\" rel=\"nofollow noopener\">The Conversation<\/a> under a Creative Commons license. Read the <a rel=\"nofollow\" target=\"_blank\" href=\"https:\/\/theconversation.com\/deepfake-detection-improves-when-using-algorithms-that-are-more-aware-of-demographic-diversity-226061\" target=\"_blank\" rel=\"nofollow noopener\">original article<\/a>.<\/em><\/p>\n<\/div>\n<blockquote><p><strong><span style=\"color: #ff6600;\">If you liked the article, do not forget to share it with your friends. Follow us on\u00a0<span style=\"color: #ff0000;\"><a style=\"color: #ff0000;\" href=\"https:\/\/news.google.com\/publications\/CAAqBwgKMN63nwsw68G3Aw\" target=\"_blank\" rel=\"nofollow noopener noreferrer\">Google News<\/a><\/span>\u00a0too, click on the star and choose us from your favorites.<\/span><\/strong><\/p><\/blockquote>\n<blockquote>\n<p style=\"text-align: center;\"><strong>If you want to read more like this article, you can visit our <span style=\"color: #ff9900;\"><a style=\"color: #ff9900;\" href=\"https:\/\/en.buradabiliyorum.com\/category\/technology\/\" target=\"_blank\" >Technology category.<\/a><\/span><\/strong><\/p>\n<\/blockquote>\n<p><span style=\"color: black;\"><a style=\"color: #ff9900;\" href=\"https:\/\/thenextweb.com\/news\/deepfake-detection-improves-algorithms-demographic-diversity\" target=\"_blank\" >Source<\/a><\/span><\/p>\n","protected":false},"excerpt":{"rendered":"<p>Deepfakes \u2013 essentially putting words in someone else\u2019s mouth in a very believable way \u2013 are becoming more sophisticated by the day and increasingly hard to spot. Recent examples of deepfakes include Taylor Swift nude images, an audio recording of President Joe Biden telling New Hampshire residents not to vote, and a video of Ukrainian&#8230;<\/p>\n","protected":false},"author":1,"featured_media":652368,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"fifu_image_url":"https:\/\/img-cdn.tnwcdn.com\/image\/tnw-blurple?filter_last=1&fit=1280%2C640&url=https%3A%2F%2Fcdn0.tnwcdn.com%2Fwp-content%2Fblogs.dir%2F1%2Ffiles%2F2025%2F02%2Fcash-macanaya-X9Cemmq4YjM-unsplash.jpg&signature=3af1017e9e7ac23ab5aa2c2be7fdaddc","fifu_image_alt":"","footnotes":""},"categories":[18],"tags":[],"class_list":["post-652367","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-technology"],"_links":{"self":[{"href":"https:\/\/buradabiliyorum.com\/en\/wp-json\/wp\/v2\/posts\/652367","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/buradabiliyorum.com\/en\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/buradabiliyorum.com\/en\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/buradabiliyorum.com\/en\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/buradabiliyorum.com\/en\/wp-json\/wp\/v2\/comments?post=652367"}],"version-history":[{"count":0,"href":"https:\/\/buradabiliyorum.com\/en\/wp-json\/wp\/v2\/posts\/652367\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/buradabiliyorum.com\/en\/wp-json\/wp\/v2\/media\/652368"}],"wp:attachment":[{"href":"https:\/\/buradabiliyorum.com\/en\/wp-json\/wp\/v2\/media?parent=652367"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/buradabiliyorum.com\/en\/wp-json\/wp\/v2\/categories?post=652367"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/buradabiliyorum.com\/en\/wp-json\/wp\/v2\/tags?post=652367"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}