{"628783":{"#nid":"628783","#data":{"type":"news","title":"Making Sure Computing Machines Don\u2019t Stereotype People","body":[{"value":"\u003Cp\u003EMachine learning algorithms dominate society, from helping judges with courtroom decisions to influencing banks on who gets loans. With big and small decisions potentially being swayed by these mathematical equations, research has become dedicated to making algorithms more transparent and fair.\u003C\/p\u003E\r\n\r\n\u003Cp\u003E\u003Cstrong\u003EUthaipon (Tao) Tantipongpipat\u003C\/strong\u003E and \u003Cstrong\u003ESamira Samadi,\u003C\/strong\u003E Georgia Tech Ph.D. students in the \u003Ca href=\u0022https:\/\/scs.gatech.edu\/\u0022\u003ESchool of Computer Science\u003C\/a\u003E, recently published a \u003Ca href=\u0022https:\/\/arxiv.org\/pdf\/1902.11281.pdf\u0022\u003Enew paper\u003C\/a\u003E that takes large data sets for population analysis and reduces the dimension of those data sets while also preserving essential traits of the groups being analyzed. Algorithms can handle millions of records but the process might compress information and lose details. This, in turn, can lead to groups of people being unfairly associated with certain behaviors or characteristics.\u003C\/p\u003E\r\n\r\n\u003Cp\u003ESamadi and Tantipongpipat\u0026rsquo;s \u003Ca href=\u0022https:\/\/www.cc.gatech.edu\/news\/615576\/georgia-tech-researchers-working-improve-fairness-ml-pipeline\u0022\u003Eprevious work\u003C\/a\u003E uses principal component analysis (PCA), a dimension reduction technique that has been the gold standard for analyzing large data sets more efficiently. Their own version, Fair-PCA, uses the strength of PCA and retains more information so that algorithms can, in theory, have better data for decision-making.\u003C\/p\u003E\r\n\r\n\u003Cp\u003EIn their latest work, the duo is optimizing Fair-dimensionality reduction, allowing populations to be more accurately represented when not only using PCA, but a wider class of dimension reduction techniques.\u003C\/p\u003E\r\n\r\n\u003Cp\u003EThe updated algorithm incorporates multiple equity measurements for populations \u0026ndash; i.e. with respect to social and economic welfare \u0026ndash; and takes into account multiple demographical attributes. For example, gender is usually analyzed as male and female, but this leaves transgender people and other non-binary people out of an algorithm\u0026rsquo;s calculations leading to unfair or biased assessments.\u003C\/p\u003E\r\n\r\n\u003Cp\u003EThis new work is designed to allow machine learning researchers to analyze complex data sets more accurately, potentially leading to less bias.\u003C\/p\u003E\r\n\r\n\u003Cp\u003E\u0026quot;I feel like if fairness and bias are not being taken seriously into account at this point, then our problems are only going to compound. Machine learning algorithms are dominating our lives every day and they learn to behave based on previous outcomes. If we just let this build up and if we don\u0026#39;t take care of it now, it will have a huge impact, one that may not be as positive as we had hoped,\u0026rdquo; said Samadi.\u003C\/p\u003E\r\n\r\n\u003Cp\u003EThe team will present \u003Ca href=\u0022https:\/\/arxiv.org\/pdf\/1902.11281.pdf\u0022\u003E\u003Cem\u003EMulti-Criteria Dimensionality Reduction with Applications to Fairness\u003C\/em\u003E\u003C\/a\u003E\u0026nbsp;in December at the \u003Ca href=\u0022https:\/\/neurips.cc\/\u0022\u003E33\u003Csup\u003Erd\u003C\/sup\u003E Annual Conference on Neural Information Processing Systems (NeurIPS)\u003C\/a\u003E 2019 in Vancouver, British Columbia.\u003C\/p\u003E\r\n","summary":null,"format":"limited_html"}],"field_subtitle":"","field_summary":"","field_summary_sentence":[{"value":"Georgia Tech researchers develop an algorithm that is less biased towards different populations."}],"uid":"34773","created_gmt":"2019-11-08 15:33:06","changed_gmt":"2019-11-08 16:14:41","author":"ablinder6","boilerplate_text":"","field_publication":"","field_article_url":"","dateline":{"date":"2019-11-08T00:00:00-05:00","iso_date":"2019-11-08T00:00:00-05:00","tz":"America\/New_York"},"extras":[],"hg_media":{"628782":{"id":"628782","type":"image","title":"This summer, Samira Samadi presented work at the International Conference on Machine Learning.","body":null,"created":"1573226939","gmt_created":"2019-11-08 15:28:59","changed":"1573226939","gmt_changed":"2019-11-08 15:28:59","alt":"Samira Samadi","file":{"fid":"239466","name":"-4936141608470894095_IMG_3150.jpg","image_path":"\/sites\/default\/files\/images\/-4936141608470894095_IMG_3150.jpg","image_full_path":"http:\/\/www.tlwarc.hg.gatech.edu\/\/sites\/default\/files\/images\/-4936141608470894095_IMG_3150.jpg","mime":"image\/jpeg","size":456090,"path_740":"http:\/\/www.tlwarc.hg.gatech.edu\/sites\/default\/files\/styles\/740xx_scale\/public\/images\/-4936141608470894095_IMG_3150.jpg?itok=5gQhr8mo"}}},"media_ids":["628782"],"groups":[{"id":"47223","name":"College of Computing"},{"id":"576481","name":"ML@GT"},{"id":"50875","name":"School of Computer Science"}],"categories":[{"id":"134","name":"Student and Faculty"},{"id":"8862","name":"Student Research"},{"id":"153","name":"Computer Science\/Information Technology and Security"}],"keywords":[],"core_research_areas":[{"id":"39501","name":"People and Technology"}],"news_room_topics":[],"event_categories":[],"invited_audience":[],"affiliations":[],"classification":[],"areas_of_expertise":[],"news_and_recent_appearances":[],"phone":[],"contact":[{"value":"\u003Cp\u003EAllie McFadden\u003C\/p\u003E\r\n\r\n\u003Cp\u003ECommunications Officer\u003C\/p\u003E\r\n\r\n\u003Cp\u003Eallie.mcfadden@cc.gatech.edu\u003C\/p\u003E\r\n","format":"limited_html"}],"email":[],"slides":[],"orientation":[],"userdata":""}}}