.png){kind=icon}
.png){kind=icon}
Return
Variance-reduction for variational inequality problems with Bregman distance function
.png){kind=icon}
PRE
AI
DOI：10.1080/10556788.2025.2612351.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
In this paper, we address variational inequalities (VIs) with a finite-sum structure. We introduce a novel and unified stochastic variance-reduced algorithm, utilizing the Bregman distance function, that can be applied to both monotone and non-monotone settings. We establish optimal convergence guarantees under the monotone case. For the non-monotone setting, we explore a structured class of problems that exhibit weak Minty solutions and analyse the complexity of our method, demonstrating improvements over existing approaches. Numerical experiments are provided to showcase the superior performance of our algorithm compared to state-of-the-art methods.
Keywords:
Variational inequality
variance reduction
bergman distance
Journal
O
IF:
1.4
Papers:
24
Citations:
0