Nonconvex Zeroth-Order Stochastic ADMM Methods with Lower Function Query Complexity
This addresses optimization challenges in machine learning where gradients are unavailable, offering incremental improvements for tasks like adversarial attacks on black-box neural networks.
The paper tackles the problem of high function query complexity and limited applicability of zeroth-order methods for nonconvex optimization with penalties and constraints, proposing ZO-SPIDER-ADMM and ZOO-ADMM+ methods that achieve lower query complexities, improving existing best results by factors like O(d^{1/3}n^{1/6}) and O(ε^{-1/2}).
Zeroth-order (a.k.a, derivative-free) methods are a class of effective optimization methods for solving complex machine learning problems, where gradients of the objective functions are not available or computationally prohibitive. Recently, although many zeroth-order methods have been developed, these approaches still have two main drawbacks: 1) high function query complexity; 2) not being well suitable for solving the problems with complex penalties and constraints. To address these challenging drawbacks, in this paper, we propose a class of faster zeroth-order stochastic alternating direction method of multipliers (ADMM) methods (ZO-SPIDER-ADMM) to solve the nonconvex finite-sum problems with multiple nonsmooth penalties. Moreover, we prove that the ZO-SPIDER-ADMM methods can achieve a lower function query complexity of $O(nd+dn^{\frac{1}{2}}ε^{-1})$ for finding an $ε$-stationary point, which improves the existing best nonconvex zeroth-order ADMM methods by a factor of $O(d^{\frac{1}{3}}n^{\frac{1}{6}})$, where $n$ and $d$ denote the sample size and data dimension, respectively. At the same time, we propose a class of faster zeroth-order online ADMM methods (ZOO-ADMM+) to solve the nonconvex online problems with multiple nonsmooth penalties. We also prove that the proposed ZOO-ADMM+ methods achieve a lower function query complexity of $O(dε^{-\frac{3}{2}})$, which improves the existing best result by a factor of $O(ε^{-\frac{1}{2}})$. Extensive experimental results on the structure adversarial attack on black-box deep neural networks demonstrate the efficiency of our new algorithms.