Measuring Fairness in Generative Models

2021/07/16 by Christopher T. H. Teo, Teo, Christopher T. H, Ngai‐Man Cheung +1 · 4 citations
Computer Science · Social Sciences · #Ethics and Social Impacts of AI #Explainable Artificial Intelligence (XAI) #FOS: Computer and information sciences #Generative Adversarial Networks and Image Synthesis #Machine Learning (cs.LG)

paper · pdf · doi:10.48550/arxiv.2107.07754

openalex publication_date 2021/07/16 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

Deep generative models have made much progress in improving training stability and quality of generated data. Recently there has been increased interest in the fairness of deep-generated data. Fairness is important in many applications, e.g. law enforcement, as biases will affect efficacy. Central to fair data generation are the fairness metrics for the assessment and evaluation of different generative models. In this paper, we first review fairness metrics proposed in previous works and highlight potential weaknesses. We then discuss a performance benchmark framework along with the assessment of alternative metrics.

Citations

• Auto-Encoding Variational Bayes
• Unrolled Generative Adversarial Networks
• NIPS 2016 Tutorial: Generative Adversarial Networks
• Improved Training of Wasserstein GANs
• FairGAN: Fairness-aware Generative Adversarial Networks
• Large Scale GAN Training for High Fidelity Natural Image Synthesis
• Deep Residual Learning for Image Recognition
• Deep Learning Face Attributes in the Wild
• Self-supervised GAN: Analysis and Improvement with Multi-class Minimax\n Game

Cited by

• FairImagen: Post-Processing for Bias Mitigation in Text-to-Image Models
• Rank of Matrices Arising out of Singular Kernel Functions
• Responsible Diffusion: A Comprehensive Survey on Safety, Ethics, and Trust in Diffusion Models
• On Optimal Steering to Achieve Exact Fairness

Related

• On Measuring Fairness in Generative Models
• Data vs. Model Machine Learning Fairness Testing: An Empirical Study
• Metrics and methods for a systematic comparison of fairness-aware machine learning algorithms
• Toward Valid Measurement Of (Un)fairness For Generative AI: A Proposal For Systematization Through The Lens Of Fair Equality of Chances
• A Review of Fairness and A Practical Guide to Selecting Context-Appropriate Fairness Metrics in Machine Learning