Title:
Multivariate Statistics and Machine Learning Under a Modern Optimization Lens

dc.contributor.author Bertsimas, Dimitris
dc.contributor.corporatename Georgia Institute of Technology. School of Industrial and Systems Engineering en_US
dc.contributor.corporatename Massachusetts Institute of Technology. School of Management en_US
dc.date.accessioned 2015-03-13T18:30:50Z
dc.date.available 2015-03-13T18:30:50Z
dc.date.issued 2015-03-05
dc.description Presented as part of the H. Milton Stewart School of Industrial and Systems Engineering 2015 Distinguished Scholarship Lecture on March 5, 2015 at 3:00 p.m. in the Bill Moore Student Success Center, Clary Theater. en_US
dc.description Dimitris Bertsimas is currently the Boeing Professor of Operations Research and the co-director of the Operations Research Center at MIT. He has been with the MIT faculty since 1988. His research interests include optimization, statistics and applied probability and their applications in health care, finance, operations management and transportation.
dc.description Runtime: 66:18 minutes
dc.description.abstract Key problems of classification and regression can naturally be written as optimization problems. While continuous optimization approaches has had a significant impact in statistics, discrete optimization has played a very limited role, primarily based on the belief that mixed integer optimization models are computationally intractable. While such beliefs were accurate two decades ago, the field of discrete optimization has made very substantial progress. Dr. Bertsimas will discuss how to apply modern first order optimization methods to find feasible solutions for classical problems in statistics, and mixed integer optimization to improve the solutions and to prove optimality by finding matching lower bounds. Specifically, he will report results for the classical variable selection problem in regression currently solved by LASSO heuristically, least quantile regression, and factor analysis. He will also present an approach to build regression models based on mixed integer optimization. In all cases he will demonstrate that the solutions found by modern optimization methods outperform the classical approaches. Most importantly, he suggests that the belief widely held in statistics that mixed integer optimization is not practically relevant for statistics applications needs to be revisited. en_US
dc.embargo.terms null en_US
dc.format.extent 66:18 minutes
dc.identifier.uri http://hdl.handle.net/1853/53223
dc.publisher Georgia Institute of Technology en_US
dc.subject Discrete optimization en_US
dc.subject Factor analysis en_US
dc.subject Regression models en_US
dc.subject Statistics en_US
dc.title Multivariate Statistics and Machine Learning Under a Modern Optimization Lens en_US
dc.type Moving Image
dc.type.genre Lecture
dspace.entity.type Publication
local.contributor.corporatename H. Milton Stewart School of Industrial and Systems Engineering
local.contributor.corporatename College of Engineering
relation.isOrgUnitOfPublication 29ad75f0-242d-49a7-9b3d-0ac88893323c
relation.isOrgUnitOfPublication 7c022d60-21d5-497c-b552-95e489a06569
Files
Original bundle
Now showing 1 - 3 of 3
No Thumbnail Available
Name:
bertsimas.mp4
Size:
533.14 MB
Format:
MP4 Video file
Description:
Download video
No Thumbnail Available
Name:
bertsimas_videostream.html
Size:
985 B
Format:
Hypertext Markup Language
Description:
Streaming Video
No Thumbnail Available
Name:
transcription.txt
Size:
55.52 KB
Format:
Plain Text
Description:
Transcription
License bundle
Now showing 1 - 1 of 1
No Thumbnail Available
Name:
license.txt
Size:
3.13 KB
Format:
Item-specific license agreed upon to submission
Description: