Mitigating Demographic Bias of Federated Learning Models via Robust-Fair Domain Smoothing: A Domain-Shifting Approach

Keyphrases

• Learning Model 100%
• Domain Shift 100%
• Federated Learning 100%
• Demographic Bias 100%
• Reference Domain 33%
• Data Distribution 33%
• Global Model 33%
• Global Distribution 33%
• Hurt 16%
• Performance Gain 16%
• Evaluation Results 16%
• Heterogeneous Data 16%
• Local Model 16%
• Local Distribution 16%
• Promising Solutions 16%
• Global Datasets 16%
• Machine Learning Models 16%
• Local Data 16%
• Distributed Data Sources 16%
• Unique Domain 16%
• Imbalanced Distribution 16%
• Training Framework 16%
• Domain Mismatch 16%
• Fairness-aware 16%

Computer Science

• Federated Learning 100%
• Data Distribution 50%
• Domain Reference 33%
• Evaluation Result 16%
• Performance Gain 16%
• Heterogeneous Data 16%
• Distributed Data Source 16%
• Local Distribution 16%
• Imbalanced Data 16%
• Machine Learning 16%
• Learning System 16%