Skip to yearly menu bar Skip to main content


Asymptotic Behaviors of Projected Stochastic Approximation: A Jump Diffusion Perspective

Jiadong Liang · Yuze Han · Xiang Li · Zhihua Zhang

Keywords: [ Jump Diffusion ] [ Asymptotic Biased ] [ Loopless Algorithm ] [ Diffusion Approximation ] [ Asymptotic Normal ] [ Bias-variance Tradeoff ] [ federated learning ]

Abstract: In this paper, we consider linearly constrained stochastic approximation problems with federated learning (FL) as a special case. We propose a stochastic approximation algorithm named by LPSA with probabilistic projections to ensure feasibility so that projections are performed with probability $p_n$ at the $n$-th iteration. Considering a specific family of the probability $p_n$ and step size $\eta_n$, we analyze our algorithm from an asymptotic and continuous perspective. Using a novel jump diffusion approximation, we show that the trajectories consisting of properly rescaled last iterates weakly converge to the solution of specific SDEs. By analyzing the SDEs, we identify the asymptotic behaviors of LPSA for different choices of $(p_n, \eta_n)$. We find the algorithm presents an intriguing asymptotic bias-variance trade-off according to the relative magnitude of $p_n$ w.r.t. $\eta_n$. It provides insights on how to choose appropriate $\{(p_n, \eta_n)\}_{n \geq 1}$ to minimize the projection complexity.

Chat is not available.