具有内在隐私的通信和能源高效的无线联邦学习（英文版）

英文标题：Communication and Energy Efficient Wireless Federated Learning with Intrinsic Privacy中文摘要：本文提出了一种无线边缘私有联邦学习方案 PFELS，通过奇异值稀疏化降低通信通信和能源消耗，提高交付速度，保证客户端级别的差分隐私，实现共享全局模型过程中的隐私安全保障。实验结果表明，与现有技术相比，PFELS 在保持同等差分隐私保障的同时能提高准确率和节省通信和能源成本。英文摘要：Federated Learning (FL) is a collaborative learning framework that enablesedge devices to collaboratively learn a global model while keeping raw datalocally. Although FL avoids leaking direct information from local datasets,sensitive information can still be inf