Fast Incremental Method for Nonconvex Optimization

英文摘要

We analyze a fast incremental aggregated gradient method for optimizing nonconvex problems of the form $\min_x \sum_i f_i(x)$. Specifically, we analyze the SAGA algorithm within an Incremental First-order Oracle framework, and show that it converges to a stationary point provably faster than both gradient descent and stochastic gradient descent. We also discuss a Polyak's special class of nonconvex problems for which SAGA converges at a linear rate to the global optimum. Finally, we analyze the practically valuable regularized and minibatch variants of SAGA. To our knowledge, this paper presents the first analysis of fast convergence for an incremental aggregated gradient method for nonconvex problems.

发表机构

• Carnegie Mellon University（卡内基梅隆大学）
• Massachusetts Institute of Technology（麻省理工学院）

机构由 AI 辅助整理，请以论文原文为准。