{"663688":{"#nid":"663688","#data":{"type":"event","title":"PhD Defense by Yiling Luo","body":[{"value":"\u003Cp\u003E\u003Cstrong\u003EThesis Title:\u0026nbsp;\u003C\/strong\u003EStochastic Methods in Model Estimation: New Algorithms and New Properties\u003C\/p\u003E\r\n\r\n\u003Cp\u003E\u0026nbsp;\u003C\/p\u003E\r\n\r\n\u003Cp\u003E\u003Cstrong\u003EThesis Committee:\u003C\/strong\u003E\u003C\/p\u003E\r\n\r\n\u003Cp\u003E1 Dr.\u0026nbsp;Xiaoming\u0026nbsp;Huo (Advisor, ISyE, Gatech)\u003C\/p\u003E\r\n\r\n\u003Cp\u003E2 Dr. Yajun Mei (Co-advisor, ISyE, Gatech)\u003C\/p\u003E\r\n\r\n\u003Cp\u003E3 Dr. Arkadi Nemirovski (ISyE, Gatech)\u003C\/p\u003E\r\n\r\n\u003Cp\u003E4 Dr. Vladimir Koltchinskii (Mathematics, Gatech)\u003C\/p\u003E\r\n\r\n\u003Cp\u003E5 Dr. Kai Zhang\u0026nbsp;(Department of Statistics and Operations Research,\u0026nbsp;University of North Carolina, Chapel Hill)\u003C\/p\u003E\r\n\r\n\u003Cp\u003E\u0026nbsp;\u003C\/p\u003E\r\n\r\n\u003Cp\u003E\u003Cstrong\u003EDate and Time:\u0026nbsp;\u003C\/strong\u003EWednesday,\u0026nbsp;December 21st, 11:00 am (EST)\u003C\/p\u003E\r\n\r\n\u003Cp\u003E\u0026nbsp;\u003C\/p\u003E\r\n\r\n\u003Cp\u003E\u003Cstrong\u003EIn-Person Location\u003C\/strong\u003E\u003Cstrong\u003E:\u0026nbsp;\u003C\/strong\u003EGroseclose 303\u003C\/p\u003E\r\n\r\n\u003Cp\u003E\u003Cstrong\u003EMeeting Link\u003C\/strong\u003E:\u0026nbsp;\u003Ca href=\u0022https:\/\/teams.microsoft.com\/l\/meetup-join\/19:meeting_MTZjMzA4NzgtM2Q5Yi00YzNlLWE0MzQtNWY2NTFjYWE0Yjkx@thread.v2\/0?context=%7B%22Tid%22:%22482198bb-ae7b-4b25-8b7a-6d7f32faa083%22,%22Oid%22:%220ad58c33-2d96-4990-a3ff-0910e5dc4e33%22%7D\u0022\u003EClick here to join the meeting\u003C\/a\u003E\u003C\/p\u003E\r\n\r\n\u003Cp\u003EMeeting ID: 295 937 083 365\u003C\/p\u003E\r\n\r\n\u003Cp\u003EPasscode: tXvJLq\u003C\/p\u003E\r\n\r\n\u003Cp\u003E\u0026nbsp;\u003C\/p\u003E\r\n\r\n\u003Cp\u003E\u0026nbsp;\u003C\/p\u003E\r\n\r\n\u003Cp\u003E\u003Cstrong\u003EAbstract:\u003C\/strong\u003E\u003C\/p\u003E\r\n\r\n\u003Cp\u003EWe study the properties of applying stochastic algorithms to solve optimization problems in model estimation.\u0026nbsp;In particular, we investigate the statistical properties of estimators that are based on some stochastic algorithms in Chapters 2-5; we propose a new stochastic algorithm and study its optimization property in Chapter 6.\u0026nbsp;\u003C\/p\u003E\r\n\r\n\u003Cp\u003E\u0026nbsp;\u003C\/p\u003E\r\n\r\n\u003Cp\u003EWe summarize the main contents in each chapter as follows.\u0026nbsp;\u003C\/p\u003E\r\n\r\n\u003Cp\u003E\u0026nbsp;\u003C\/p\u003E\r\n\r\n\u003Cp\u003EIn Chapter 2, we explore a directional bias phenomenon in both stochastic gradient descent and gradient descent, and examine their implications for the resulting estimators.\u0026nbsp;We would argue that the outcome from the stochastic gradient descent may lead to a better generalization error bound.\u003C\/p\u003E\r\n\r\n\u003Cp\u003E\u0026nbsp;\u003C\/p\u003E\r\n\r\n\u003Cp\u003EIn Chapter 3, we study a property of implicit regularization by a variance reduction version of the stochastic mirror descent algorithm.\u0026nbsp;The phenomenon of implicit regularization by applying certain algorithms has attracted a lot of attention, and its existence with the variance reduction based stochastic algorithm is new.\u0026nbsp;\u003C\/p\u003E\r\n\r\n\u003Cp\u003E\u0026nbsp;\u003C\/p\u003E\r\n\r\n\u003Cp\u003EIn Chapter 4, we establish the equivalence between the variance reduced stochastic mirror descents with a technique that has been developed in information theory -- variance reduced stochastic natural gradient descent.\u0026nbsp;The purpose of establishing such an equivalence is that the properties of both problems can automatically be shared with each other.\u0026nbsp;\u003C\/p\u003E\r\n\r\n\u003Cp\u003E\u0026nbsp;\u003C\/p\u003E\r\n\r\n\u003Cp\u003EIn Chapter 5, we study a recent algorithm -- ROOT-SGD -- for the online learning problem, and we estimate the covariance of the estimator that is computed via the ROOT-SGD algorithm.\u0026nbsp;Our covariance estimators quantify the uncertainty in the ROOT-SGD algorithm, which are useful for statistical inference.\u0026nbsp;\u003C\/p\u003E\r\n\r\n\u003Cp\u003E\u0026nbsp;\u003C\/p\u003E\r\n\r\n\u003Cp\u003EIn Chapter 6, we study a constrained strongly convex problem -- the entropic OT, and we propose a primal-dual stochastic algorithm with variance reduction to solve it.\u0026nbsp;We show that the computational complexity of our algorithm is better than other first-order algorithms for solving the entropic OT.\u0026nbsp;\u003C\/p\u003E\r\n\r\n\u003Cp\u003E\u0026nbsp;\u003C\/p\u003E\r\n","summary":null,"format":"limited_html"}],"field_subtitle":"","field_summary":"","field_summary_sentence":[{"value":"Stochastic Methods in Model Estimation: New Algorithms and New Properties"}],"uid":"27707","created_gmt":"2022-12-07 13:40:54","changed_gmt":"2022-12-07 13:40:54","author":"Tatianna Richardson","boilerplate_text":"","field_publication":"","field_article_url":"","field_event_time":{"event_time_start":"2022-12-21T11:00:00-05:00","event_time_end":"2022-12-21T13:00:00-05:00","event_time_end_last":"2022-12-21T13:00:00-05:00","gmt_time_start":"2022-12-21 16:00:00","gmt_time_end":"2022-12-21 18:00:00","gmt_time_end_last":"2022-12-21 18:00:00","rrule":null,"timezone":"America\/New_York"},"extras":[],"groups":[{"id":"221981","name":"Graduate Studies"}],"categories":[],"keywords":[{"id":"100811","name":"Phd Defense"}],"core_research_areas":[],"news_room_topics":[],"event_categories":[{"id":"1788","name":"Other\/Miscellaneous"}],"invited_audience":[{"id":"78761","name":"Faculty\/Staff"},{"id":"78771","name":"Public"},{"id":"78751","name":"Undergraduate students"}],"affiliations":[],"classification":[],"areas_of_expertise":[],"news_and_recent_appearances":[],"phone":[],"contact":[],"email":[],"slides":[],"orientation":[],"userdata":""}}}