$\bar{G}_{mst}$:An Unbiased Stratified Statistic and a Fast Gradient Optimization Algorithm Based on It

no code implementations • 7 Oct 2021 • Aixiang Chen

-The fluctuation effect of gradient expectation and variance caused by parameter update between consecutive iterations is neglected or confusing by current mainstream gradient optimization algorithms.

A Novel Stochastic Stratified Average Gradient Method: Convergence Rate and Its Complexity

no code implementations • 21 Oct 2017 • Aixiang Chen, Bingchuan Chen, Xiaolong Chai, Rui Bian, Hengguang Li

SGD (Stochastic Gradient Descent) is a popular algorithm for large scale optimization problems due to its low iterative cost.