On the convergence of stochastic forward- backward-forward algorithms with variance reduction in pseudo-monotone variational inequalities
On the convergence of stochastic forward- backward-forward algorithms with variance reduction in pseudo-monotone variational inequalities
We develop a new stochastic algorithm with variance reduction for solving pseudomonotone stochastic variational inequalities. Our method builds on Tseng’s forward-backward-forward algorithm, which is known in the deterministic literature to be a valuable alternative to Korpelevich’s extragradient method when solving variational inequalities over a convex and closed set governed with pseudomonotone and Lipschitz continuous operators. The main computational advantage of Tseng’s algorithm is that it relies only on a single projection step, and two independent queries of a stochastic oracle. Our algorithm incorporates a variance reduction mechanism, and leads to a.s. convergence to solutions of a merely pseudo-monotone stochastic variational inequality problem. To the best of our knowledge, this is the first stochastic algorithm achieving this by using only a single projection at each iteration.
1-5
Staudigl, Mathias
4556db35-f650-4b0d-b933-39c116a5933a
Bot, Radu Ioan
d4eaeb1e-c774-4c45-a273-b078e66b2af7
Phan, Tu
52577e5d-ebe9-4a43-b5e7-68aa06cfdcaf
Mertikopoulos, Panayotis
5bd29601-0511-4eac-90b8-079bf7ed5b76
7 December 2018
Staudigl, Mathias
4556db35-f650-4b0d-b933-39c116a5933a
Bot, Radu Ioan
d4eaeb1e-c774-4c45-a273-b078e66b2af7
Phan, Tu
52577e5d-ebe9-4a43-b5e7-68aa06cfdcaf
Mertikopoulos, Panayotis
5bd29601-0511-4eac-90b8-079bf7ed5b76
Staudigl, Mathias, Bot, Radu Ioan, Phan, Tu and Mertikopoulos, Panayotis
(2018)
On the convergence of stochastic forward- backward-forward algorithms with variance reduction in pseudo-monotone variational inequalities.
NIPS 2018 Workshop on Smooth Games, Optimization and Machine Learning.
07 Dec 2018.
.
Record type:
Conference or Workshop Item
(Paper)
Abstract
We develop a new stochastic algorithm with variance reduction for solving pseudomonotone stochastic variational inequalities. Our method builds on Tseng’s forward-backward-forward algorithm, which is known in the deterministic literature to be a valuable alternative to Korpelevich’s extragradient method when solving variational inequalities over a convex and closed set governed with pseudomonotone and Lipschitz continuous operators. The main computational advantage of Tseng’s algorithm is that it relies only on a single projection step, and two independent queries of a stochastic oracle. Our algorithm incorporates a variance reduction mechanism, and leads to a.s. convergence to solutions of a merely pseudo-monotone stochastic variational inequality problem. To the best of our knowledge, this is the first stochastic algorithm achieving this by using only a single projection at each iteration.
Text
FBF-NIPS
- Accepted Manuscript
More information
Published date: 7 December 2018
Venue - Dates:
NIPS 2018 Workshop on Smooth Games, Optimization and Machine Learning, 2018-12-07 - 2018-12-07
Identifiers
Local EPrints ID: 434610
URI: http://eprints.soton.ac.uk/id/eprint/434610
PURE UUID: c02e9a20-f39b-43e7-8ff9-8e749e5de4e2
Catalogue record
Date deposited: 03 Oct 2019 16:30
Last modified: 16 Mar 2024 04:42
Export record
Contributors
Author:
Mathias Staudigl
Author:
Radu Ioan Bot
Author:
Panayotis Mertikopoulos
Download statistics
Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.
View more statistics