Online Learning with Multiple Fairness Regularizers via Graph-Structured Feedback

TMLR Paper6264 Authors

Abstract: There is an increasing need to enforce multiple, often competing, measures of fairness within automated decision systems. The appropriate weighting of these fairness objectives is typically unknown a priori, may change over time and, in our setting, must be learned adaptively through sequential interactions. In this work, we address this challenge in a bandit setting, where decisions are made with graph-structured feedback.

Submission Type: Regular submission (no more than 12 pages of main content)

Changes Since Last Submission: In this revision, we have: (1) rewritten the abstract, intro, and conclusion to clarify contributions; (2) strengthened experiments with a new baseline and more trials; and (3) added a notation table and corrected errors throughout. Some typos are corrected.

Assigned Action Editor: ~Ian_A._Kash1

Submission Number: 6264