Multiple Fairness Metric

Multiple fairness metrics in machine learning aim to address the complex and multifaceted nature of algorithmic bias by evaluating models across various fairness criteria, rather than relying on a single metric. Current research focuses on developing methods to efficiently compute and compare multiple fairness metrics simultaneously, often employing techniques like multi-objective optimization and reinforcement learning to improve model fairness without sacrificing utility. This work is crucial for ensuring responsible AI development, as it allows for a more comprehensive assessment of bias and facilitates the creation of fairer and more equitable algorithms across diverse applications. The ultimate goal is to move beyond simplistic notions of fairness towards a more nuanced understanding and mitigation of bias in real-world systems.

Papers

June 25, 2024

Fairpriori: Improving Biased Subgroup Discovery for Deep Neural Network Fairness
Kacy Zhou, Jiawen Wen, Nan Yang, Dong Yuan, Qinghua Lu, Huaming Chen
Neural Network Group Fairness Fairness Metric Intersectional Bias Subgroup Discovery FAIrness Testing Multiple Fairness Metric

December 7, 2023

A Brief Tutorial on Sample Size Calculations for Fairness Audits
Harvineet Singh, Fan Xia, Mi-Ok Kim, Romain Pirracchio, Rumi Chunara, Jean Feng
Fairness Metric Sample Size Fairness Audit Multiple Fairness Metric

November 9, 2023

Kantian Deontology Meets AI Alignment: Towards Morally Grounded Fairness Metrics
Carlos Mougan, Joshua Brand
AI Alignment AI Fairness Moral Theory Multiple Fairness Metric

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

June 19, 2023

Fairness Index Measures to Evaluate Bias in Biometric Recognition
Ketan Kotwal, Sebastien Marcel
Absolute Stance Bias Biometric Identification Fairness Assessment Fairness Measure Demographic Fairness Multiple Fairness Metric

February 8, 2023

On the Richness of Calibration
Benedikt Höltgen, Robert C Williamson
Calibration Performance Algorithmic Fairness Fairness Notion Generating Rich Multiple Fairness Metric

February 5, 2023

The Unfairness of Fair Machine Learning: Levelling down and strict egalitarianism by default
Brent Mittelstadt, Sandra Wachter, Chris Russell
Procedural Fairness Fairness Constraint Fair Machine Learning Fairness Measure Egalitarian Fairness Multiple Fairness Metric Default Policy

October 30, 2022

Mitigating Unfairness via Evolutionary Multi-objective Ensemble Learning
Zhang Qingquan, Liu Jialin, Zhang Zeqi, Wen Junyi, Mao Bifei, Yao Xin
Fairness Metric Fairness Measure Fairness Mitigation Multiple Fairness Metric

February 10, 2022

Fair When Trained, Unfair When Deployed: Observable Fairness Measures are Unstable in Performative Prediction Settings
Alan Mishler, Niccolò Dalmasso
Distribution Shift Fairness Notion Performative Prediction Deodata Predictor Multiple Fairness Metric

Multiple Fairness Metric

Papers

Fairpriori: Improving Biased Subgroup Discovery for Deep Neural Network Fairness

A Brief Tutorial on Sample Size Calculations for Fairness Audits

Kantian Deontology Meets AI Alignment: Towards Morally Grounded Fairness Metrics

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

Fairness Index Measures to Evaluate Bias in Biometric Recognition

On the Richness of Calibration

The Unfairness of Fair Machine Learning: Levelling down and strict egalitarianism by default

Mitigating Unfairness via Evolutionary Multi-objective Ensemble Learning

Fair When Trained, Unfair When Deployed: Observable Fairness Measures are Unstable in Performative Prediction Settings