BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:Sequential Fair Allocation: Achieving the Optimal Envy-Efficiency 
 Tradeoff Curve
DTSTART:20221207T093000
DTEND:20221207T103000
DTSTAMP:20260406T204334Z
UID:f9411777c96fdd9b076a11b84182f406bf899ee487f5ee3b17dc6a80
CATEGORIES:Conferences - Seminars
DESCRIPTION:Dr. Sean Sinclair\nAbstract\nOptimizing the operations of comp
 lex systems often involves making tradeoffs between objectives such as eff
 iciency\, revenue\, and fairness. How these criteria interact is often not
  well understood\, and current approaches focus on maximizing a convex com
 bination which provides little operational insights. In this talk we inves
 tigate the tradeoff between fairness and efficiency in online resource all
 ocation motivated by a partnership with the Food Bank of the Southern Tier
 . We start by establishing an uncertainty principle: a lower bound exactly
  characterizing the envy and efficiency Pareto frontier. We complement thi
 s by showing how to leverage the principle of algorithmic guardrails\, art
 ificial constraints imposed on the set of actions\, which allows algorithm
 s to exactly match the uncertainty principle. These techniques extend to a
  variety of settings including perishable resources\, evolving budgets\, a
 nd a wide range of individual preference models.\nThis work falls under a 
 broader range of questions in designing practical sequential decision maki
 ng algorithms for uncertain environments. Such questions include optimizin
 g multiple objectives\, as discussed above\, but also in designing algorit
 hms which computationally and statistically scale to real-world systems. T
 ime permitting\, I will highlight this by briefly discussing my work desig
 ning algorithms which leverage information relaxation in problems with exo
 genous dynamics. The algorithm appeals to existing computational solutions
  for business problems for solving large-scale deterministic optimization 
 problems. This algorithm design is currently in deployment for the Microso
 ft Azure platform.\nBiography\nSean Sinclair is a fifth-year Ph.D. candida
 te in Operations Research and Information Engineering at Cornell Universit
 y\, co advised by Siddhartha Banerjee and Christina Yu. He received his un
 dergraduate degree in Honours Mathematics and Computer Science from McGill
  University and afterwards served as a teacher in Ghana with the Peace Cor
 ps. His research focuses on developing algorithms for data-driven sequenti
 al decision making in societal systems.\nSean was selected for the 2022 Fu
 ture Leaders Summit at the Michigan Institute for Data Science. In 2020 an
 d 2022 respectively he was a visitor at the Simons Institute for the progr
 ams on the Theory of Reinforcement Learning and Data-Driven Decision Proce
 sses. During the Summer of 2021 he was a research intern at Microsoft Rese
 arch in the Reinforcement Learning group working on virtual machine schedu
 ling in Microsoft Azure.
LOCATION:ODY 4 03 https://plan.epfl.ch/?room==ODY%204%2003 https://epfl.zo
 om.us/j/67366997671?pwd=bzlzSEgyOGxtdURDTTIrRytFZUNmZz09
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR