FedCiR: Client-Invariant Representation Learning for Federated Non-IID Features

Keyphrases

• Mutual Information 100%
• Global Representation 100%
• Non-IID 100%
• Invariant Representation Learning 100%
• Data Distribution 66%
• Federated Learning 66%
• Learning Paradigm 33%
• Local Data 33%
• Learning Framework 33%
• Global Model 33%
• Invariant Features 33%
• Data-driven Model 33%
• Training Accuracy 33%
• Invariant Feature Extraction 33%
• Regularizer 33%
• Distributed Learning 33%
• Edge Devices 33%
• Federated Learning System 33%
• Generalization Error Bound 33%
• Data Heterogeneity 33%
• Data Meaning 33%
• Approximate Distribution 33%
• Non-independent and Identically Distributed Data 33%
• Training Convergence 33%
• Source-free 33%

Computer Science

• Federated Learning 100%
• Invariant Representation 100%
• Representation Learning 100%
• Mutual Information 100%
• Learning Framework 66%
• Data Distribution 66%
• Generalization Error 33%
• Data Heterogeneity 33%
• Approximation (Algorithm) 33%
• Distributed Learning 33%
• Feature Extraction 33%