PEFed: Enhancing privacy and efficiency in federated learning via removable perturbation and decentralized encryption

Keyphrases

• Encryption 100%
• Federated Learning 100%
• Privacy-preserving 40%
• Classification Task 40%
• Regression Task 40%
• Client-side Encryption 40%
• Secure Aggregation 40%
• Model Update 20%
• Machine Learning 20%
• Classification Accuracy 20%
• Computational Cost 20%
• Learning Methods 20%
• Secure multi-party Computation 20%
• Local Data 20%
• Success Rate 20%
• Learning Approaches 20%
• Medical Dataset 20%
• Parallel Algorithm 20%
• Loss Function 20%
• Data Processing Methods 20%
• Superior Accuracy 20%
• Privacy-preserving Machine Learning 20%
• Perturbation Scheme 20%
• Single Instruction multiple Data 20%
• Distributed Processing 20%
• Inference Attack 20%
• Global Gradient 20%
• Diverse Datasets 20%
• Perturbation Vector 20%
• Data Reconstruction Attack 20%
• Gradient Preservation 20%

Computer Science

• Federated machine learning 100%
• Privacy Preserving 33%
• Classification Task 33%
• Regression Task 33%
• Classification Accuracy 16%
• Machine Learning 16%
• Learning System 16%
• secure multiparty computation 16%
• Computational Cost 16%
• Privacy Preservation 16%
• Processing Method 16%
• Learning Approach 16%
• Data Processing 16%
• Single Instruction Multiple Data 16%