Workshop: OPT2020: Optimization for Machine Learning

Contributed Video: Can We Find Near-Approximately-Stationary Points of Nonsmooth Nonconvex Functions?, Ohad Shamir

Ohad Shamir

Abstract: It is well-known that given a bounded, smooth nonconvex function, standard gradient-based methods can find $\epsilon$-stationary points (where the gradient norm is less than $\epsilon$) in $O(1/\epsilon^2)$ iterations. However, many important nonconvex optimization problems, such as those associated with training modern neural networks, are inherently \emph{not} smooth, making these results inapplicable. Moreover, as recently pointed out in \citet{zhang2020complexity}, it is generally impossible to provide finite-time guarantees for finding an $\epsilon$-stationary point of nonsmooth functions. Perhaps the most natural relaxation of this is to find points which are *near* such $\epsilon$-stationary points. In this paper, we show that even this relaxed goal is hard to obtain in general, given only black-box access to the function values and gradients. We also discuss the pros and cons of alternative approaches.