BEGIN:VCALENDAR
VERSION:2.0
PRODID:icalendar-ruby
CALSCALE:GREGORIAN
X-WR-CALNAME:ORIE Colloquium: Ken Moon (Cornell SC Johnson College of Busin
 ess)
X-WR-TIMEZONE:Eastern Time (US & Canada)
BEGIN:VEVENT
DTSTAMP:20260512T055747Z
UID:tag:localist.com\,2008:EventInstance_51817871943827
DTSTART:20260224T211500Z
DTEND:20260224T221500Z
DESCRIPTION:(Machine) Learning Preferences from Complex Choice Sets: An App
 lication to Service Networks\n\nHidden complexity pervades commonplace cho
 ices. For example\, customers at everyday venues such as supermarkets\, sh
 opping centers\, and amusement parks routinely choose from the combinatori
 ally many available paths through all or some of the venues' stations (e.g
 .\, sections\, stores\, rides). Rich data now capture such customer choice
 s\, but the complexity\, e.g.\, a factorial rate of growth of paths in ven
 ue size\, remains a challenge for empirical or empirically grounded modeli
 ng research. Firstly\, it is unclear how customers assess such complex dec
 isions. Secondly\, choice estimation methods exhibit fundamental issues ra
 nging from intractability to inconsistency when estimating consumer prefer
 ences from such data. Under commonplace conditions\, we show that a hidden
  but provably low-dimensional fraction of relevant choice comparisons suff
 ices to capture all customer preference information from data and can be a
 utomatically learned and exploited by machine learning (ML). The same low 
 dimensionality further justifies simple shopping heuristics for customers.
  We develop a neural network-based estimator of customer preferences and d
 emonstrate it on data tracking shoppers at a hypermarket serving over 1M c
 ustomers annually. Such estimation uncovers the hypermarket’s network de
 mand structure\, and modest capacity reallocations counterfactually raise 
 equilibrium service throughput by 5-25% through. Addressing a long-standin
 g challenge in marketing and operations\, we identify complementary statio
 ns from congestion levels at one or more stations producing cross-station 
 demand effects at others.\n \n\nBio: Ken Moon is an associate professor of
  operations\, technology and information management at the Cornell SC John
 son College of Business. He uses large datasets to study and optimize the 
 organization of workers and of markets. His work applies mathematical mode
 ling\, causal analysis\, and algorithms to improve the performance and tre
 atment of workforces\, operations of online markets\, and design of polici
 es and regulations for complex networks. His recent interest is in utilizi
 ng machine learning methods in advanced inference. He received his Ph.D. f
 rom the Stanford Graduate School of Business\, a J.D. from Harvard Law Sch
 ool\, and an undergraduate degree from Stanford University.
GEO:42.443451;-76.481506
LOCATION:Frank H. T. Rhodes Hall\, 253
SUMMARY:ORIE Colloquium: Ken Moon (Cornell SC Johnson College of Business)
URL;VALUE=URI:https://events.cornell.edu/event/orie-colloquium-ken-moon-cor
 nell-sc-johnson-college-of-business
CATEGORIES:Colloquium
END:VEVENT
END:VCALENDAR