Proportionally Fair Clustering

The Data Science Discipline of the School of EECS is hosting the following guest seminar:

Proportionally Fair Clustering 

Speaker: Professor Haris Aziz, UNSW Sydney

Abstract:

In recent years, there has been a surge in effort to formalize notions of fairness in machine learning. We focus on centroid clustering---one of the fundamental tasks in unsupervised machine  learning. We propose a new axiom ``proportionally representative fairness'' (PRF) that is designed for clustering problems where the selection of centroids  reflects the distribution of data points and how tightly they are clustered together. Our fairness concept is not satisfied by existing fair clustering algorithms. We design efficient algorithms to achieve PRF both for unconstrained and discrete clustering problems. Our algorithm for the unconstrained setting is also the first known polynomial-time approximation algorithm for the well-studied Proportional Fairness (PF) axiom. Our algorithm for the discrete setting also matches the best known approximation factor for PF. We then discuss how the ideas have important implications for the problem of fair transit stop placement where agents travel between source–destination pairs and may either walk directly or utilize a shuttle service via selected transit stops. 

Based on joint work with Barton E. Lee, Sean Morota Chu, and Jeremy Vollen.

Bio:

Haris Aziz is a Professor at UNSW Sydney and leader of the Algorithmic Decision Theory group. He was the founding director of the UNSW AI Institute, one of the largest AI institutes with over 300 faculty members. His research interests lie at the intersection of artificial intelligence, theoretical computer science and mathematical social sciences ---, especially computational social choice and algorithmic game theory. Haris is a recipient of the Scientia Fellowship (2018 - 2024), CORE Chris Wallace Research Excellence Award (2017) and the Julius Career Award (2016 - 2018). In 2016, he was selected by the Institute of Electrical and Electronics Engineers (IEEE) for the AI 10 to Watch List. He is the editor of TheoretiCS and an associate editor of several journals including Artificial Intelligence, ACM Transactions on Economics and Computation, Autonomous Agents and Multi-Agent Systems, and Social Choice & Welfare. He was on the board of directors of International Foundation for Autonomous Agents and Multiagent Systems (IFAAMAS) and was the chair of the IFAAMAS Awards Committee.