Efficient Federated Learning With Enhanced Privacy via Lottery Ticket Pruning in Edge Computing
Yifan Shi, Kang Wei, Li Shen, Jun Li, Xueqian Wang, Bo Yuan, Song Guo
Abstract
Federated learning (FL) can train collaboratively with several mobile terminals (MTs), which faces critical challenges in communication, resource, and privacy. Existing privacy-preserving methods usually adopt instance-level differential privacy (DP), which provides a rigorous privacy guarantee but with several bottlenecks: performance degradation, transmission overhead, and resource constraints. Therefore, we propose Fed-LTP, an efficient and privacy-enhanced FL framework with <underline xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><b>L</b></u> ottery <underline xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><b>T</b></u> icket <underline xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><b>H</b></u> ypothesis (LTH) and zero-concentrated D <underline xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><b>P</b></u> (zCDP). It generates a pruned global model on the server side and conducts sparse-to-sparse training from scratch with zCDP on the client side. On the server side, two pruning schemes are proposed: (i) the weight-based pruning (LTH) determines the pruned global model structure; (ii) the iterative pruning further shrinks the size of the pruned model. Meanwhile, the performance of Fed-LTP is boosted via model validation based on the Laplace mechanism. On the client side, we use sparse-to-sparse training to solve the resource-constraints issue and provide tighter privacy analysis to reduce the privacy budget. We evaluate the effectiveness of Fed-LTP on several real-world datasets in both independent and identically distributed (IID) and non-IID settings. The results confirm the superiority of Fed-LTP over state-of-the-art (SOTA) methods in communication, computation, and memory efficiencies while realizing a better utility-privacy trade-off.