New Insight into Hybrid Stochastic Gradient Descent: Beyond With-Replacement Sampling and Convexity

Pan Zhou, Xiaotong Yuan, Jiashi Feng

Neural Information Processing Systems 

As an incremental-gradient algorithm, the hybrid stochastic gradient descent (HS-GD) enjoys merits of both stochastic and full gradient methods for finite-sum problem optimization.

Similar Docs  Excel Report  more

TitleSimilaritySource
None found