{"id":159409,"date":"2010-06-01T00:00:00","date_gmt":"2010-06-01T00:00:00","guid":{"rendered":"https:\/\/www.microsoft.com\/en-us\/research\/msr-research-item\/web-scale-bayesian-click-through-rate-prediction-for-sponsored-search-advertising-in-microsofts-bing-search-engine\/"},"modified":"2018-10-16T21:57:32","modified_gmt":"2018-10-17T04:57:32","slug":"web-scale-bayesian-click-through-rate-prediction-for-sponsored-search-advertising-in-microsofts-bing-search-engine","status":"publish","type":"msr-research-item","link":"https:\/\/www.microsoft.com\/en-us\/research\/publication\/web-scale-bayesian-click-through-rate-prediction-for-sponsored-search-advertising-in-microsofts-bing-search-engine\/","title":{"rendered":"Web-Scale Bayesian Click-Through Rate Prediction for Sponsored Search Advertising in Microsoft&#8217;s Bing Search Engine"},"content":{"rendered":"<div class=\"asset-content\">\n<p>We describe a new Bayesian click-through rate (CTR) prediction algorithm used for Sponsored Search in Microsoft\u2019s Bing search engine. The algorithm is based on a probit regression model that maps discrete or real-valued input features to probabilities. It maintains Gaussian beliefs over weights of the model and performs Gaussian online updates derived from approximate message passing. Scalability of the algorithm is ensured through a principled weight pruning procedure and an approximate parallel implementation. We discuss the challenges arising from evaluating and tuning the predictor as part of the complex system of sponsored search where the predictions made by the algorithm decide about future training sample composition. Finally, we show experimental results from the production system and compare to a calibrated Na\u00efve Bayes algorithm.<\/p>\n<\/div>\n<p><!-- .asset-content --><\/p>\n","protected":false},"excerpt":{"rendered":"<p>We describe a new Bayesian click-through rate (CTR) prediction algorithm used for Sponsored Search in Microsoft\u2019s Bing search engine. The algorithm is based on a probit regression model that maps discrete or real-valued input features to probabilities. It maintains Gaussian beliefs over weights of the model and performs Gaussian online updates derived from approximate message [&hellip;]<\/p>\n","protected":false},"featured_media":0,"template":"","meta":{"msr-url-field":"","msr-podcast-episode":"","msrModifiedDate":"","msrModifiedDateEnabled":false,"ep_exclude_from_search":false,"_classifai_error":"","footnotes":""},"msr-content-type":[3],"msr-research-highlight":[],"research-area":[13556,13555],"msr-publication-type":[193716],"msr-product-type":[],"msr-focus-area":[],"msr-platform":[],"msr-download-source":[],"msr-locale":[268875],"msr-post-option":[],"msr-field-of-study":[],"msr-conference":[],"msr-journal":[],"msr-impact-theme":[],"msr-pillar":[],"class_list":["post-159409","msr-research-item","type-msr-research-item","status-publish","hentry","msr-research-area-artificial-intelligence","msr-research-area-search-information-retrieval","msr-locale-en_us"],"msr_publishername":"","msr_edition":"Proceedings of the 27th International Conference on Machine Learning ICML 2010, Invited Applications Track (unreviewed, to appear)","msr_affiliation":"","msr_published_date":"2010-06-01","msr_host":"","msr_duration":"","msr_version":"","msr_speaker":"","msr_other_contributors":"","msr_booktitle":"Proceedings of the 27th International Conference on Machine Learning ICML 2010, Invited Applications Track (unreviewed, to appear)","msr_pages_string":"","msr_chapter":"","msr_isbn":"","msr_journal":"","msr_volume":"","msr_number":"","msr_editors":"","msr_series":"","msr_issue":"","msr_organization":"","msr_how_published":"","msr_notes":"Invited Applications Track","msr_highlight_text":"","msr_release_tracker_id":"","msr_original_fields_of_study":"","msr_download_urls":"","msr_external_url":"","msr_secondary_video_url":"","msr_longbiography":"","msr_microsoftintellectualproperty":1,"msr_main_download":"221449","msr_publicationurl":"","msr_doi":"","msr_publication_uploader":[{"type":"file","title":"AdPredictor ICML 2010 &#8211; final.pdf","viewUrl":"https:\/\/www.microsoft.com\/en-us\/research\/wp-content\/uploads\/2010\/06\/AdPredictor-ICML-2010-final.pdf","id":221449,"label_id":0}],"msr_related_uploader":"","msr_attachments":[{"id":221449,"url":"https:\/\/www.microsoft.com\/en-us\/research\/wp-content\/uploads\/2010\/06\/AdPredictor-ICML-2010-final.pdf"}],"msr-author-ordering":[{"type":"user_nicename","value":"thoreg","user_id":34034,"rest_url":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/microsoft-research\/v1\/researchers?person=thoreg"},{"type":"user_nicename","value":"joaquinc","user_id":32347,"rest_url":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/microsoft-research\/v1\/researchers?person=joaquinc"},{"type":"user_nicename","value":"tborcher","user_id":33903,"rest_url":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/microsoft-research\/v1\/researchers?person=tborcher"},{"type":"user_nicename","value":"rherb","user_id":33390,"rest_url":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/microsoft-research\/v1\/researchers?person=rherb"}],"msr_impact_theme":[],"msr_research_lab":[],"msr_event":[],"msr_group":[],"msr_project":[169873],"publication":[],"video":[],"download":[],"msr_publication_type":"inproceedings","related_content":{"projects":[{"ID":169873,"post_title":"TrueSkill\u2122 Ranking System","post_name":"trueskill-ranking-system","post_type":"msr-project","post_date":"2005-11-18 07:02:09","post_modified":"2024-06-20 05:32:08","post_status":"publish","permalink":"https:\/\/www.microsoft.com\/en-us\/research\/project\/trueskill-ranking-system\/","post_excerpt":"The TrueSkill ranking system is a skill based ranking system for\u00a0Xbox Live developed at Microsoft Research. The purpose of a ranking system is to both identify and track the skills of gamers in a game (mode) in order to be able to match them into competitive matches. TrueSkill has been used to rank and match players in many different games, from Halo 3 to Forza Motorsport 7. An improved version of the TrueSkill ranking system,&hellip;","_links":{"self":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-project\/169873"}]}}]},"_links":{"self":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-research-item\/159409"}],"collection":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-research-item"}],"about":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/types\/msr-research-item"}],"version-history":[{"count":2,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-research-item\/159409\/revisions"}],"predecessor-version":[{"id":540650,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-research-item\/159409\/revisions\/540650"}],"wp:attachment":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/media?parent=159409"}],"wp:term":[{"taxonomy":"msr-content-type","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-content-type?post=159409"},{"taxonomy":"msr-research-highlight","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-research-highlight?post=159409"},{"taxonomy":"msr-research-area","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/research-area?post=159409"},{"taxonomy":"msr-publication-type","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-publication-type?post=159409"},{"taxonomy":"msr-product-type","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-product-type?post=159409"},{"taxonomy":"msr-focus-area","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-focus-area?post=159409"},{"taxonomy":"msr-platform","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-platform?post=159409"},{"taxonomy":"msr-download-source","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-download-source?post=159409"},{"taxonomy":"msr-locale","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-locale?post=159409"},{"taxonomy":"msr-post-option","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-post-option?post=159409"},{"taxonomy":"msr-field-of-study","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-field-of-study?post=159409"},{"taxonomy":"msr-conference","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-conference?post=159409"},{"taxonomy":"msr-journal","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-journal?post=159409"},{"taxonomy":"msr-impact-theme","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-impact-theme?post=159409"},{"taxonomy":"msr-pillar","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-pillar?post=159409"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}