Adaptive Stochastic Variance Reduction for Non-convex Finite-Sum Minimization

Ali Kavis · Stratis Skoulakis · Kimon Antonakopoulos · Leello Tadesse Dadi · Volkan Cevher

Hall J #816

Keywords: [ variance reduction ] [ nonconvex optimization ] [ Adaptive Methods ] [ finite-sum minimization ]

Abstract: We propose an adaptive variance-reduction method, called AdaSpider, for minimization of $L$-smooth, non-convex functions with a finite-sum structure. In essence, AdaSpider combines an AdaGrad-inspired (Duchi et al., 2011), but a fairly distinct, adaptive step-size schedule with the recursive \textit{stochastic path integrated estimator} proposed in (Fang et al., 2018). To our knowledge, AdaSpider is the first parameter-free non-convex variance-reduction method in the sense that it does not require the knowledge of problem-dependent parameters, such as smoothness constant $L$, target accuracy $\epsilon$ or any bound on gradient norms. In doing so, we are able to compute an $\epsilon$-stationary point with $\tilde{O}\left(n + \sqrt{n}/\epsilon^2\right)$ oracle-calls, which matches the respective lower bound up to logarithmic factors.

Chat is not available.