{"650239":{"#nid":"650239","#data":{"type":"event","title":"ISyE Department Seminar - Dmitriy Drusvyatskiy","body":[{"value":"\u003Ch3\u003E\u003Cstrong\u003ETitle:\u003C\/strong\u003E\u003C\/h3\u003E\r\n\r\n\u003Cp\u003EStochastic optimization under distributional shifts\u003C\/p\u003E\r\n\r\n\u003Ch3\u003E\u003Cstrong\u003EAbstract:\u003C\/strong\u003E\u003C\/h3\u003E\r\n\r\n\u003Cp\u003ELearning problems commonly exhibit an interesting feedback\u003Cbr \/\u003E\r\nmechanism wherein the population data reacts to decision makers\u0026#39;\u003Cbr \/\u003E\r\nactions. This is the case for example when members of the population\u003Cbr \/\u003E\r\nrespond to a deployed classifier by manipulating their features so as\u003Cbr \/\u003E\r\nto improve the likelihood of being positively labeled. In this way,\u003Cbr \/\u003E\r\nthe population is manipulating the learning process by distorting the\u003Cbr \/\u003E\r\ndata distribution that is accessible to the\u0026nbsp;learner. In this talk, I will\u0026nbsp;present some recent modelling frameworks and algorithms for dynamic\u0026nbsp;problems of this type, rooted in stochastic optimization and game\u0026nbsp;theory.\u003Cbr \/\u003E\r\n\u003Cbr \/\u003E\r\nJoint work with Evan Faulkner (UW), Maryam Fazel (UW), Adhyyan Narang\u003Cbr \/\u003E\r\n(UW), Lillian J. Ratliff (UW), Lin Xiao (Facebook AI)\u003C\/p\u003E\r\n\r\n\u003Ch3\u003E\u003Cstrong\u003EBio:\u003C\/strong\u003E\u003C\/h3\u003E\r\n\r\n\u003Cp\u003EDmitriy Drusvyatskiy received his PhD from the Operations\u003Cbr \/\u003E\r\nResearch and Information Engineering department at Cornell University\u003Cbr \/\u003E\r\nin 2013, followed by a post doctoral appointment in the Combinatorics\u003Cbr \/\u003E\r\nand Optimization department at Waterloo, 2013-2014. He joined the\u003Cbr \/\u003E\r\nMathematics department at University of Washington as an Assistant\u003Cbr \/\u003E\r\nProfessor in 2014, and was promoted to an Associate Professor in 2019.\u003Cbr \/\u003E\r\nDmitriy\u0026#39;s research broadly focuses on designing and analyzing\u003Cbr \/\u003E\r\nalgorithms for large-scale optimization problems, primarily motivated\u003Cbr \/\u003E\r\nby applications in data science. Dmitriy has received a number of\u003Cbr \/\u003E\r\nawards, including the Air Force Office of Scientific Research (AFOSR)\u003Cbr \/\u003E\r\nYoung Investigator Program (YIP) Award, NSF CAREER, INFORMS\u003Cbr \/\u003E\r\nOptimization Society Young Researcher Prize 2019, and finalist\u003Cbr \/\u003E\r\ncitations for the Tucker Prize 2015 and the Young Researcher Best\u003Cbr \/\u003E\r\nPaper Prize at ICCOPT 2019. Dmitriy is currently a co-PI of the NSF\u003Cbr \/\u003E\r\nfunded Transdisciplinary Research in Principles of Data Science\u003Cbr \/\u003E\r\n(TRIPODS) institute at University of Washington.\u003Cbr \/\u003E\r\n\u003Cbr \/\u003E\r\nResearch currently supported by NSF CAREER DMS 1651851 and NSF CCF 1740551.\u003C\/p\u003E\r\n","summary":null,"format":"limited_html"}],"field_subtitle":"","field_summary":[{"value":"\u003Ch3\u003E\u003Cstrong\u003EAbstract:\u003C\/strong\u003E\u0026nbsp;\u003C\/h3\u003E\r\n\r\n\u003Cp\u003ELearning problems commonly exhibit an interesting feedback\u003Cbr \/\u003E\r\nmechanism wherein the population data reacts to decision makers\u0026#39;\u003Cbr \/\u003E\r\nactions. This is the case for example when members of the population\u003Cbr \/\u003E\r\nrespond to a deployed classifier by manipulating their features so as\u003Cbr \/\u003E\r\nto improve the likelihood of being positively labeled. In this way,\u003Cbr \/\u003E\r\nthe population is manipulating the learning process by distorting the\u003Cbr \/\u003E\r\ndata distribution that is accessible to the\u0026nbsp;learner. In this talk, I will\u003C\/p\u003E\r\n\r\n\u003Cp\u003Epresent some recent modelling frameworks and algorithms for dynamic\u003Cbr \/\u003E\r\nproblems of this type, rooted in stochastic optimization and game\u003Cbr \/\u003E\r\ntheory.\u003Cbr \/\u003E\r\n\u003Cbr \/\u003E\r\nJoint work with Evan Faulkner (UW), Maryam Fazel (UW), Adhyyan Narang\u003Cbr \/\u003E\r\n(UW), Lillian J. Ratliff (UW), Lin Xiao (Facebook AI)\u003Cbr \/\u003E\r\n\u0026nbsp;\u003C\/p\u003E\r\n","format":"limited_html"}],"field_summary_sentence":[{"value":"Stochastic optimization under distributional shifts"}],"uid":"34868","created_gmt":"2021-08-30 19:55:22","changed_gmt":"2021-11-16 17:43:07","author":"sbryantturner3","boilerplate_text":"","field_publication":"","field_article_url":"","field_event_time":{"event_time_start":"2021-11-19T11:00:00-05:00","event_time_end":"2021-11-19T12:00:00-05:00","event_time_end_last":"2021-11-19T12:00:00-05:00","gmt_time_start":"2021-11-19 16:00:00","gmt_time_end":"2021-11-19 17:00:00","gmt_time_end_last":"2021-11-19 17:00:00","rrule":null,"timezone":"America\/New_York"},"extras":[],"groups":[{"id":"1242","name":"School of Industrial and Systems Engineering (ISYE)"}],"categories":[],"keywords":[],"core_research_areas":[],"news_room_topics":[],"event_categories":[{"id":"1795","name":"Seminar\/Lecture\/Colloquium"}],"invited_audience":[{"id":"78761","name":"Faculty\/Staff"},{"id":"177814","name":"Postdoc"},{"id":"78771","name":"Public"},{"id":"174045","name":"Graduate students"},{"id":"78751","name":"Undergraduate students"}],"affiliations":[],"classification":[],"areas_of_expertise":[],"news_and_recent_appearances":[],"phone":[],"contact":[],"email":[],"slides":[],"orientation":[],"userdata":""}}}