Learning Fair Classifiers via Min-Max F-divergence Regularization

Computer Science

• Machine Learning 100%
• Regularization 100%
• Decision-Making 50%
• Postprocessing 50%
• Art Performance 50%
• Classification Models 50%
• Preprocessing 50%
• Multiple Domain 50%
• Regularization Method 50%
• Sensitive Attribute 50%
• Fairness Constraint 50%

Keyphrases

• F-divergence 100%
• Fair Classification 40%
• Multiple Sensitive Attributes 20%
• Gradient-based Training 20%
• Group Fairness 20%
• Equalized Odds 20%
• COMPAS 20%
• Convexity Properties 20%
• CelebA Dataset 20%
• F-divergence Measure 20%
• Aided Decision 20%
• Differentiability Properties 20%
• Variational Representation 20%