From Pseudorandomness to Multi-Group Fairness and Back

TL;DR

This paper explores the connections between multi-group fairness in algorithms and pseudorandomness concepts, introducing new variants of multicalibration, efficient algorithms, and graph-theoretic results.

Contribution

It introduces new variants of multicalibration based on statistical distance, linking fairness and pseudorandomness, and provides novel algorithms and theoretical results.

Findings

New variants of multicalibration based on statistical distance

More efficient algorithms for multicalibration

Graph-theoretic results and a hardcore lemma for real-valued functions

Abstract

We identify and explore connections between the recent literature on multi-group fairness for prediction algorithms and the pseudorandomness notions of leakage-resilience and graph regularity. We frame our investigation using new variants of multicalibration based on statistical distance and closely related to the concept of outcome indistinguishability. Adopting this perspective leads us not only to new, more efficient algorithms for multicalibration, but also to our graph theoretic results and a proof of a novel hardcore lemma for real-valued functions.