FairGAN: Fairness-aware Generative Adversarial Networks
arXiv:1805.11202
Abstract
Fairness-aware learning is increasingly important in data mining. Discrimination prevention aims to prevent discrimination in the training data before it is used to conduct predictive analysis. In this paper, we focus on fair data generation that ensures the generated data is discrimination free. Inspired by generative adversarial networks (GAN), we present fairness-aware generative adversarial networks, called FairGAN, which are able to learn a generator producing fair data and also preserving good data utility. Compared with the naive fair data generation models, FairGAN further ensures the classifiers which are trained on generated data can achieve fair classification on real data. Experiments on a real dataset show the effectiveness of FairGAN.
References in corpus (9)
- Conditional Generative Adversarial Nets
- Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks
- Equality of Opportunity in Supervised Learning
- Fairness in Machine Learning: Lessons from Political Philosophy
- Generating Multi-label Discrete Patient Records using Generative Adversarial Networks
- Data Decisions and Theoretical Implications when Adversarially Learning Fair Representations
- Learning Adversarially Fair and Transferable Representations
- Censoring Representations with an Adversary
- Fairness Through Awareness
Cited by in corpus (10)
- A Review on Generative Adversarial Networks: Algorithms, Theory, and Applications
- Fairness for Robust Log Loss Classification
- Improved Adversarial Learning for Fair Classification
- DECAF: Generating Fair Synthetic Data Using Causally-Aware Generative Networks
- Adversarial training approach for local data debiasing
- Analyzing Fairness of Computer Vision and Natural Language Processing Models
- Measuring Fairness in Generative Models
- FairWASP: Fast and Optimal Fair Wasserstein Pre-processing
- Additive Adversarial Learning for Unbiased Authentication
- Slice Tuner: A Selective Data Acquisition Framework for Accurate and Fair Machine Learning Models