Zeroth-Order Nonconvex Nonsmooth Optimization with Heavy-Tailed Noise

ArXi:2605.24513v1 Announce Type: new This paper considers the nonconvex nonsmooth problem in which the objective function is Lipschitz continuous. We focus on the stochastic setting where the algorithm can access stochastic function value evaluations with heavy-tailed noise, which is prevalent in many popular machine learning applications. We propose a stochastic zeroth-order algorithm that refines the framework of online-to-nonconvex conversion by clipping the two-point gradient estimator.