首页> 外文会议>Annual Allerton Conference on Communication, Control, and Computing >Sequential smoothing framework for convex-concave saddle point problems with application to large-scale constrained optimization
【24h】

Sequential smoothing framework for convex-concave saddle point problems with application to large-scale constrained optimization

机译:凸凹鞍点问题的顺序平滑框架及其在大规模约束优化中的应用

获取原文

摘要

In this paper, we propose a sequential smoothing algorithm for solving large scale convex-concave saddle point problems. We prove that this class of algorithms achieves the convergence rate O(1/√ε) in the deterministic setting and O(1/√ε) with probability at least 1 - δ, where δ ϵ (0, 1), in the stochastic setting, to obtain an g-optimal solution. We then apply our general algorithm to specific problems in large-scale convex constrained optimization.
机译:在本文中,我们提出了一种解决大型凸凹鞍点问题的顺序平滑算法。我们证明这类算法在确定性设置中达到了收敛速度O(1 /√ε),并且在概率上达到O(1 /√ε)的概率至少为1-δ,其中δϵ(0,1)是随机的设置,以获得g最优解。然后,我们将通用算法应用于大规模凸约束优化中的特定问题。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号