Fairness Utility Trade

Fairness-utility trade-offs in machine learning explore the inherent tension between maximizing a model's predictive accuracy (utility) and ensuring its predictions are fair across different demographic groups. Current research focuses on developing methods to quantify and navigate this trade-off, often employing Pareto efficiency to identify optimal solutions balancing both objectives, and utilizing techniques like reinforcement learning and bi-objective loss functions to achieve this balance. This research is crucial for mitigating algorithmic bias in high-stakes applications, leading to more equitable and trustworthy AI systems across various domains.

Papers

November 13, 2024

TowerDebias: A Novel Debiasing Method based on the Tower Property
Norman Matloff, Aditya Mittal
Human Prediction Procedural Fairness Black Box Model Self Debiasing Debiasing Framework Black Box Machine Learning Model Fairness Utility Trade

April 15, 2024

Utility-Fairness Trade-Offs and How to Find Them
Sepehr Dehdashtian, Bashir Sadeghi, Vishnu Naresh Boddeti
Fairness Performance Trade Demographic Fairness Fairness Utility Trade Fair Utility

February 22, 2024

February 15, 2024

DPBalance: Efficient and Fair Privacy Budget Scheduling for Federated Learning as a Service
Yu Liu, Zibo Wang, Yifei Zhu, Chen Chen
High Efficiency Service Provider Privacy Budget Fairness Utility Trade

July 21, 2023

Towards Better Fairness-Utility Trade-off: A Comprehensive Measurement-Based Reinforcement Learning Framework
Simiao Zhang, Jitao Bai, Menghong Guan, Yihao Huang, Yueling Zhang, Jun Sun, Geguang Pu
Fairness Notion Reinforcement Learning Framework Multiple Fairness Metric Fairness Utility Trade

July 12, 2023

FAIRO: Fairness-aware Adaptation in Sequential-Decision Making for Human-in-the-Loop Systems
Tianyu Zhao, Mojtaba Taherisadr, Salma Elmalaki
Human in the Loop Achieving Fairness Fairness Utility Trade

April 7, 2023

Fairness through Aleatoric Uncertainty
Anique Tahir, Lu Cheng, Huan Liu
Procedural Fairness Fair Machine Learning Aleatoric Uncertainty Fairness Loss Fairness Utility Trade

June 6, 2022

A Justice-Based Framework for the Analysis of Algorithmic Fairness-Utility Trade-Offs
Corinna Hertweck, Joachim Baumann, Michele Loi, Eleonora Viganò, Christoph Heitz
General Analysis Pareto Optimal Value Laden Choice Fairness Evaluation Decision Making System Prediction Based Decision Making Fairness Utility Trade

May 27, 2022

Generalized Reductions: Making any Hierarchical Clustering Fair and Balanced with Low Cost
Marina Knittel, Max Springer, John P. Dickerson, MohammadTaghi Hajiaghayi
Hierarchical Clustering Constructive Reduction Low Cost Fair Clustering Balanced Approach Fairness Utility Trade

Fairness Utility Trade

Papers

TowerDebias: A Novel Debiasing Method based on the Tower Property

Utility-Fairness Trade-Offs and How to Find Them

Federated Fairness without Access to Sensitive Groups

Towards Efficient Pareto-optimal Utility-Fairness between Groups in Repeated Rankings

DPBalance: Efficient and Fair Privacy Budget Scheduling for Federated Learning as a Service

Towards Better Fairness-Utility Trade-off: A Comprehensive Measurement-Based Reinforcement Learning Framework

FAIRO: Fairness-aware Adaptation in Sequential-Decision Making for Human-in-the-Loop Systems

Fairness through Aleatoric Uncertainty

A Justice-Based Framework for the Analysis of Algorithmic Fairness-Utility Trade-Offs

Generalized Reductions: Making any Hierarchical Clustering Fair and Balanced with Low Cost