A Randomized Block-Coordinate Primal-Dual Method for Large-scale Stochastic Saddle Point Problems

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Hamedani, Erfan Yazdandoost, Jalilzadeh, Afrooz, Aybat, Necdet Serhat
Format: Preprint
Veröffentlicht: 2019
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866909810688000000
author Hamedani, Erfan Yazdandoost
Jalilzadeh, Afrooz
Aybat, Necdet Serhat
author_facet Hamedani, Erfan Yazdandoost
Jalilzadeh, Afrooz
Aybat, Necdet Serhat
contents We consider (stochastic) convex-concave saddle point (SP) problems with high-dimensional decision variables, arising in various applications including machine learning problems. To contend with the challenges in computing full gradients, we employ a randomized block-coordinate primal-dual scheme in which randomly selected primal and dual blocks of variables are updated. We consider both deterministic and stochastic settings, where deterministic partial gradients and their randomly sampled estimates are used, respectively, at each iteration. We investigate the convergence of the proposed method under different blocking strategies and provide the corresponding complexity results. While the best-known computational complexity result for computing a saddle point with $\varepsilon$ primal-dual gap for deterministic primal-dual methods using full gradients is $\mathcal O(\max\{m,n\}^2/\varepsilon)$, where $m$ and $n$ denote the dimensions of primal and dual variables, respectively, we show that our proposed randomized block-coordinate method achieves an improved complexity of $\mathcal O(mn/\varepsilon)$ assuming a coordinate-friendly structure on the problem. Moreover, for the stochastic setting where a mini-batch sample gradient is utilized, we show a computational complexity of $\tilde{\mathcal{O}}(m^2n^2/\varepsilon^2)$ through acceleration. Finally, almost sure convergence of the iterate sequence to a saddle point is established.
format Preprint
id arxiv_https___arxiv_org_abs_1907_03886
institution arXiv
publishDate 2019
record_format arxiv
spellingShingle A Randomized Block-Coordinate Primal-Dual Method for Large-scale Stochastic Saddle Point Problems
Hamedani, Erfan Yazdandoost
Jalilzadeh, Afrooz
Aybat, Necdet Serhat
Optimization and Control
We consider (stochastic) convex-concave saddle point (SP) problems with high-dimensional decision variables, arising in various applications including machine learning problems. To contend with the challenges in computing full gradients, we employ a randomized block-coordinate primal-dual scheme in which randomly selected primal and dual blocks of variables are updated. We consider both deterministic and stochastic settings, where deterministic partial gradients and their randomly sampled estimates are used, respectively, at each iteration. We investigate the convergence of the proposed method under different blocking strategies and provide the corresponding complexity results. While the best-known computational complexity result for computing a saddle point with $\varepsilon$ primal-dual gap for deterministic primal-dual methods using full gradients is $\mathcal O(\max\{m,n\}^2/\varepsilon)$, where $m$ and $n$ denote the dimensions of primal and dual variables, respectively, we show that our proposed randomized block-coordinate method achieves an improved complexity of $\mathcal O(mn/\varepsilon)$ assuming a coordinate-friendly structure on the problem. Moreover, for the stochastic setting where a mini-batch sample gradient is utilized, we show a computational complexity of $\tilde{\mathcal{O}}(m^2n^2/\varepsilon^2)$ through acceleration. Finally, almost sure convergence of the iterate sequence to a saddle point is established.
title A Randomized Block-Coordinate Primal-Dual Method for Large-scale Stochastic Saddle Point Problems
topic Optimization and Control
url https://arxiv.org/abs/1907.03886