Abstract: Peer review is the backbone of academic publishing. A crucial condition of proper peer review is that submitted papers are reviewed by experts in the subject area. High overall matching accuracy, referred to as "efficiency," ensures conferences publish quality and original work. Accurate reviewing is also very important to individual authors, since reviewers provide invaluable feedback and impact career outcomes. Thus reviewer assignment must maximize efficiency while ensuring no individual papers receive poorly matched reviewers. In this talk, I present two reviewer assignment algorithms that are both efficient and fair. One algorithm gives a 1.834 approximation to the maximum Nash Welfare solution, a solution which is generally considered to be both efficient and fair. The other approach aims to satisfy the "envy-free up to 1 item (EF1)" fairness criterion, while obtaining a constant factor approximation to the best utilitarian social welfare, another standard efficiency measure. Both algorithms achieve the goals of efficiency and fairness in practice, but I will discuss some important theoretical limits to the second approach which we aim to resolve in the future.
The CICS Theory Seminar is online, free and open to the public. If you are interested in giving a talk, please email Professor Immerman or Rik Sengupta. Note that in addition to being a public lecture series, this is also a one-credit graduate seminar (CompSci 891M) that can be taken repeatedly for credit.