Enabling Efficient Federated Learning via Ternary Quantization over Fading Channels

Keyphrases

• Machine Learning Models 100%
• Fading Channels 100%
• Federated Learning 100%
• Ternary Quantization 100%
• Doping Ratio 80%
• Downlink Transmission 40%
• Local Machine 40%
• Uplink Transmission 40%
• Gradient Backpropagation 40%
• Computational Efficiency 20%
• Fusion Center 20%
• Channel Condition 20%
• Channeling Effect 20%
• Additive Noise 20%
• Channel Fading 20%
• Communication Efficiency 20%
• System Resistance 20%
• Rayleigh Fading 20%
• Learning Performance 20%
• Federated Learning System 20%
• High Doping 20%
• Stochastic Gradient Descent with Momentum 20%
• User Data Privacy 20%
• Additive Channel 20%
• Fading Affect 20%
• Additive Noise Channels 20%

Computer Science

• Fading Channel 100%
• Federated Learning 100%
• Learning System 100%
• Machine Learning 83%
• Gradient Descent 50%
• Downlink Transmission 33%
• Uplink Transmission 33%
• Channel Condition 16%
• User Data 16%
• Data Privacy 16%
• Learning Performance 16%
• Rayleigh Fading 16%
• Stochastic Algorithm 16%

Engineering

• Fading Channel 100%
• Learning System 100%
• Gradient Descent 50%
• Additive Noise 33%
• Downlink and Uplink Transmission 33%
• Model Parameter 16%
• System Resistance 16%
• Channel Condition 16%
• Rayleigh Fading 16%
• Data Privacy 16%