On the linear convergence of the projected stochastic gradient method with constant step-size

By Volkan Cevher, Bang Công Vu. 2017

Subfields: Optimization & Training, Mathematical Optimization

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Optimization & Training

1. Peter Richtárik
2. Nicolas Loizou
3. Robert Mansel Gower
4. Filip Hanzely
5. Xun Qian
6. Simon Lacoste-Julien
7. Ahmed Khaled
8. Egor Shulgin

In Mathematical Optimization

1. Nicolas Loizou
2. Robert Mansel Gower
3. Xun Qian
4. Alibek Sailanbayev
5. Peter Richtárik
6. Egor Shulgin
7. Martin Takác
8. Jakub Konecný

Citations in the corpus

• SGD: General Analysis and Improved Rates (2019, background): “Based on a strong growth condition ( c 1 = 0 ), Schmidt & Roux (2013) were the first to establish linear convergence of SGD, with Cevher & Vu (2017) later giving sufficient and necessary conditions for the linear convergence of SGD under this condition.”