SPADE-DFL: Communication-Efficient Decentralized Federated Learning via Derivative-Free Linearized ADMM
SPADE-DFL is developed, a primal--dual method that allows the number of local function-value updates between neighbor exchanges to grow with the computation budget while preserving the nonprivate convergence order.
Meng-Li Wei, Meng-Kai Zhu, Jia-Wen Chen et al.
· 0 citations