Title: Fair Lotteries for the Committee Voting Problem

Speaker: Dr Mashbat Suzuki

Summary:
In the committee voting problem, we are tasked with selecting a subset of alternatives (or candidates) of a fixed size given the preferences of a population of voters. The main challenge of the problem is to aggregate the preferences of the voters in a fair and proportional manner. Due to its prevalent nature, the problem has found applications in recommender systems, the design of Q&A platforms, blockchain protocols, and global optimization. In this work, we introduce ex-ante proportional fairness properties in this domain and formalize a hierarchy of fairness notions. We also provide polynomial-time algorithms that output lotteries (probability distributions) over committees that satisfy desirable fairness properties both ex-ante and ex-post. We will then show how some of these results can be extended to the participatory budgeting setting.

Biography:
Mashbat Suzuki is a Postdoctoral Fellow in the School of Computer Science and Engineering at UNSW Sydney. Prior to joining UNSW, he finished his PhD in computer science at McGill University under the supervision of Prof. Adrian Vetta. His research interests lie at the interface between theoretical computer science and economics. In particular, his recent research has been focused on fair division and social choice.

MEETING ID: 873 1557 5255; PASSWORD: 778635

WED 17 APRIL 2024 16:00-17:00 (AEDT, Melbourne Time)