核心概念
Individual fairness auditing schemes in online learning aim to balance predictive accuracy with fairness by detecting violations and minimizing regret.
統計資料
"Using our generalized framework, we present an oracle-efficient algorithm achieving an upper bound of O(√T) for regret and O(T3/4) for the number of fairness violations."
"Our algorithms greatly reduce the computational complexity of previous approaches."
引述
"Similar individuals should be treated similarly."
"Our algorithms greatly reduce the computational complexity of previous approaches."