An analysis of stochastic variance reduced gradient for linear inverse problems
Keyword(s):
Abstract Stochastic variance reduced gradient (SVRG) is a popular variance reduction technique for stochastic gradient descent (SGD). We provide a first analysis of the method for solving a class of linear inverse problems in the lens of the classical regularization theory. We prove that for a suitable constant step size schedule, the method can achieve an optimal convergence rate in terms of the noise level (under suitable regularity condition) and the variance of the SVRG iterate error is smaller than that by SGD. These theoretical findings are corroborated by a set of numerical experiments.
2021 ◽
Vol 9
(4)
◽
pp. 1553-1588
2019 ◽
Vol 33
◽
pp. 3943-3950
2020 ◽
Vol 34
(06)
◽
pp. 10126-10135
2020 ◽
Vol 34
(04)
◽
pp. 5503-5510
2018 ◽
Vol 9
(3)
◽
pp. 24-38
Keyword(s):
1998 ◽
Vol 35
(02)
◽
pp. 395-406
◽
Wireless Brain Wave Classification for Alzheimer’s Patients via Efficient Neural Network Computation
2018 ◽
Vol 10
(03)
◽
pp. 1850004