Complexity Lower Bounds for Nonconvex-Strongly-Concave Min-Max Optimization

Abstract
We provide a first-order oracle complexity lower bound for finding stationary points of min-max optimization problems where the objective function is smooth, nonconvex in the minimization variable, and strongly concave in the maximization variable. We establish a lower bound of for deterministic oracles, where defines the level of approximate stationarity and is the condition number. Our analysis shows that the upper bound achieved in (Lin et al., 2020b) is optimal in the and dependence up to logarithmic factors. For stochastic oracles, we provide a lower bound of . It suggests that there is a significant gap between the upper bound in (Lin et al., 2020a) and our lower bound in the condition number dependence.
View on arXivComments on this paper