From Resource Allocation to Machine Learning: Fairness Through Computation and Fairness in Computation
Bhaskar Ray Chaudhury
University of Illinois Urbana-Champaign, USA
CIS@MPG Tenure-Track Faculty
Bhaskar Ray Chaudhury is currently a Future Faculty Fellow at the University of Illinois at Urbana Champaign. He received his PhD from Max Planck Institute for Informatics under the supervision of Kurt Mehlhorn and Karl Bringmann. He is broadly interested in economics and computation, computational social choice theory, and machine learning. His work on computational social choice was recognized by the Exemplary Paper in the Theory Track Award and the Best Paper with a Student Lead Author Award at the 21st ACM conference on Economics and Computation (EC 2020). He is also the recipient of the "Teachers Ranked Excellent by Students" award at UIUC.
AG 1, AG 2, AG 3, INET, AG 4, AG 5, D6, SWS, RG1, MMCI
From Plato's Republica to John Rawl's Theory of Justice, every major work on the ethical foundations of human society, has held fairness at its zenith. Today, in the age of algorithms, we are equipped with substantial computational resources, and rigorous data driven decision making processes, to address the fairness concerns better than ever (fairness through computation). Conversely, with algorithms used extensively to make decisions of large societal impact, fairness has also evolved to be an integral requirement of several algorithmic paradigms (fairness in computation). In this talk, I discuss my work pushing the frontiers on both.
(i) Fairness through computation: We give algorithmic solutions to two fundamental fairness problems in game theory.
(ii) Fairness in Computation: We describe efficient algorithms integrating fairness notions and guarantees from social choice theory, and microeconomics to fairness demanding settings in machine learning like federated learning and fairness aware classification.
Bhaskar is a CIS@MPG tenure-track faculty candidate hosted by Prof. Dr. Danupon Na Nongkai.