BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:Learning Optimal and Fair Policies for Online Allocation of Scarce
  Societal Resources from Data Collected in Deployment
DTSTART:20231004T160000
DTEND:20231004T170000
DTSTAMP:20260916T063153Z
UID:92f35410ba271298c11f0c55209fb2715bc850f0c11bbdd5571b5cfe
CATEGORIES:Conferences - Seminars
DESCRIPTION:Professor Çağıl Koçyiğit\nAbstract: We study the problem 
 of allocating scarce societal resources of different types (e.g.\, permane
 nt housing\, deceased donor kidneys for transplantation\, ventilators) to 
 heterogeneous allocatees on a waitlist (e.g.\, people experiencing homeles
 sness\, individuals suffering from end-stage renal disease\, Covid-19 pati
 ents) based on their observed covariates. We leverage administrative data 
 collected in deployment to design an online policy that maximizes expected
  outcomes while satisfying budget constraints\, in the long run. Our propo
 sed policy waitlists each individual for the resource maximizing the diffe
 rence between their estimated mean treatment outcome and the estimated res
 ource dual-price or\, roughly\, the opportunity cost of using the resource
 . Resources are then allocated as they arrive\, in a first-come first-serv
 e fashion. We demonstrate that our data-driven policy almost surely asympt
 otically achieves the expected outcome of the optimal out-of-sample policy
  under mild technical assumptions. We extend our framework to incorporate 
 various fairness constraints. We evaluate the performance of our approach 
 on the problem of designing policies for allocating scarce housing resourc
 es to people experiencing homelessness in Los Angeles based on data from t
 he homeless management information system. In particular\, we show that us
 ing our policies improves rates of exit from homelessness by 1.9% and that
  policies that are fair in either allocation or outcomes by race come at a
  very low price of fairness.\n \nJoint work with Bill Tang\, Eric Rice\, 
 and Phebe Vayanos\n\nBio sketch: Çağıl Koçyiğit is an assistant profe
 ssor at the Luxembourg Centre for Logistics and Supply Chain Management\,
  University of Luxembourg. Her research focuses on optimization under unc
 ertainty with applications in economics\, resource allocation\, and mechan
 ism design. Prior to joining the University of Luxembourg\, she received a
  Ph.D. in Management of Technology from EPFL in 2020\, where she worked wi
 th Daniel Kuhn.\n\n 
LOCATION:ODY 4 03 https://plan.epfl.ch/?room==ODY%204%2003 https://epfl.zo
 om.us/j/62066253762
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR
