Date: Monday, November 24th, 2025
10:00 am – 11:00 am Pacific Time
1:00 pm – 2:00 pm Eastern Time
Location: Weekly Seminar, Zoom
Title: Envy-Efficiency Tradeoffs in Dynamic Fair Allocation
Abstract:
In many societal settings, decision-makers are tasked with allocating a scarce resource in a way that balances fairness and efficiency. To handle these competing objectives, practitioners often resort to ad-hoc rules of thumb or optimize a weighted combination. The tradeoff between these two desiderata, on the other hand, is often not well understood or difficult to formally characterize. In this talk we introduce a line a work that seeks to provide tight characterizations of envy-efficiency tradeoffs in dynamic fair allocation for a variety of practically motivated settings. We first consider a classical setting in which the decision-maker is faced with a stochastic number of arrivals to which they must allocated a fixed budget of resources throughout the horizon, and illustrate the impact of demand uncertainty on the envy-efficiency tradeoff. However, this ignores a practical reality of many real-world settings, such as food banks, in which resources are perishable. We demonstrate that perishability significantly modifies the envy-efficiency trade-off in this latter setting, and requires the development of new algorithmic tools. Simulations calibrated to a real-world dataset demonstrate the robustness of our algorithmic framework across a variety of perishing regimes.
Bio:
Sean Sinclair is an Assistant Professor at Northwestern University in the Industrial Engineering and Management Sciences department. Previously he was a postdoctoral associate under Devavrat Shah and Ali Jadbabaie at Massachusetts Institute of Technology. Sean completed his PhD at Cornell University coadvised by Christina Lee Yu and Siddhartha Banerjee. His research focuses on developing algorithms for data-driven sequential decision making in societal applications.