{"id":775777,"date":"2021-07-07T16:54:01","date_gmt":"2021-07-07T23:54:01","guid":{"rendered":"https:\/\/www.microsoft.com\/en-us\/research\/?post_type=msr-research-item&#038;p=775777"},"modified":"2021-10-23T16:54:07","modified_gmt":"2021-10-23T23:54:07","slug":"a-truly-unbiased-model","status":"publish","type":"msr-video","link":"https:\/\/www.microsoft.com\/en-us\/research\/video\/a-truly-unbiased-model\/","title":{"rendered":"A Truly Unbiased Model"},"content":{"rendered":"<p>Any dataset has bias, as the data collection will be intervened by human bias and affected by the nature\u2019s biased law. Therefore, any machine learning model based on it will be biased and limited in generalization. Of course, we have a plenty of \u201cdebiasing\u201d techniques that deliberately remove the bias of the \u201cmajority\u201d, so as to take care of the \u201cminority\u201d &#8212; fairness is believed to be achieved. However, such \u201ctaking care of minority\u201d per se is a bias unfair to \u201cmajority\u201d. In this talk, I will share a \u201cbest of two worlds\u201d model that is truly unbiased. Its design is simple, yet effective and theoretically grounded.<\/p>\n<div>\n\t<a\n\t\thref=\"https:\/\/www.microsoft.com\/en-us\/research\/uploads\/prod\/2021\/10\/070721_Hanwang_Zhang_A-Truly-Unbiased-Model.pdf\"\n\t\tclass=\"button cta-link\"\n\t\tdata-bi-type=\"button\"\n\t\tdata-bi-cN=\"View slides\"\n\t\tdata-bi-tN=\"shortcodes\/msr-button\"\n\t\ttarget=\"_blank\" rel=\"noopener noreferrer\">\n\t\tView slides\t<\/a>\n\n\t<\/div>\n","protected":false},"excerpt":{"rendered":"<p>Any dataset has bias, as the data collection will be intervened by human bias and affected by the nature\u2019s biased law. Therefore, any machine learning model based on it will be biased and limited in generalization. Of course, we have a plenty of \u201cdebiasing\u201d techniques that deliberately remove the bias of the \u201cmajority\u201d, so as [&hellip;]<\/p>\n","protected":false},"featured_media":775780,"template":"","meta":{"msr-url-field":"","msr-podcast-episode":"","msrModifiedDate":"","msrModifiedDateEnabled":false,"ep_exclude_from_search":false,"_classifai_error":"","footnotes":""},"research-area":[13556],"msr-video-type":[259633],"msr-locale":[268875],"msr-post-option":[],"msr-impact-theme":[],"msr-pillar":[],"class_list":["post-775777","msr-video","type-msr-video","status-publish","has-post-thumbnail","hentry","msr-research-area-artificial-intelligence","msr-video-type-vision-language-summer-talk-series","msr-locale-en_us"],"msr_download_urls":"","msr_external_url":"https:\/\/youtu.be\/XnjPssqK7xA","msr_secondary_video_url":"","msr_video_file":"","_links":{"self":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-video\/775777"}],"collection":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-video"}],"about":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/types\/msr-video"}],"version-history":[{"count":3,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-video\/775777\/revisions"}],"predecessor-version":[{"id":787690,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-video\/775777\/revisions\/787690"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/media\/775780"}],"wp:attachment":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/media?parent=775777"}],"wp:term":[{"taxonomy":"msr-research-area","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/research-area?post=775777"},{"taxonomy":"msr-video-type","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-video-type?post=775777"},{"taxonomy":"msr-locale","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-locale?post=775777"},{"taxonomy":"msr-post-option","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-post-option?post=775777"},{"taxonomy":"msr-impact-theme","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-impact-theme?post=775777"},{"taxonomy":"msr-pillar","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-pillar?post=775777"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}