First-Order Algorithms for Robust Optimization Problems via Convex-Concave Saddle-Point Lagrangian Reformulation

被引:0
|
作者
Postek, Krzysztof [1 ]
Shtern, Shimrit [1 ]
机构
[1] Technion Israel Inst Technol, Fac Data & Decis Sci, IL-3200003 Haifa, Israel
基金
以色列科学基金会;
关键词
convergence analysis; first order methods; robust optimization; saddle point; VARIATIONAL-INEQUALITIES;
D O I
10.1287/ijoc.2022.0200
中图分类号
TP39 [计算机的应用];
学科分类号
081203 ; 0835 ;
摘要
Robust optimization (RO) is one of the key paradigms for solving optimization problems affected by uncertainty. Two principal approaches for RO, the robust counterpart method and the adversarial approach, potentially lead to excessively large optimization problems. For that reason, first-order approaches, based on online convex optimization, have been proposed as alternatives for the case of large-scale problems. However, existing first-order methods are either stochastic in nature or involve a binary search for the optimal value. We show that this problem can also be solved with deterministic first-order algorithms based on a saddle-point Lagrangian reformulation that avoid both of these issues. Our approach recovers the other approaches' O(1 / epsilon(2)) convergence rate in the general case and offers an improved O(1 / epsilon) rate for problems with constraints that are affine both in the decision and in the uncertainty. Experiment involving robust quadratic optimization demonstrates the numerical benefits of our approach.
引用
收藏
页数:26
相关论文
共 50 条
  • [31] Robustness of Accelerated First-Order Algorithms for Strongly Convex Optimization Problems
    Mohammadi, Hesameddin
    Razaviyayn, Meisam
    Jovanovic, Mihailo R.
    IEEE TRANSACTIONS ON AUTOMATIC CONTROL, 2021, 66 (06) : 2480 - 2495
  • [32] Distributed Coordination for Nonsmooth Convex Optimization via Saddle-Point Dynamics
    Cortes, Jorge
    Niederlaender, Simon K.
    JOURNAL OF NONLINEAR SCIENCE, 2019, 29 (04) : 1247 - 1272
  • [33] On Robust Saddle-Point Criterion in Optimization Problems with Curvilinear Integral Functionals
    Treanta, Savin
    Das, Koushik
    MATHEMATICS, 2021, 9 (15)
  • [34] A Second Order Primal-Dual Dynamical System for a Convex-Concave Bilinear Saddle Point Problem
    He, Xin
    Hu, Rong
    Fang, Yaping
    APPLIED MATHEMATICS AND OPTIMIZATION, 2024, 89 (02):
  • [35] Customized proximal point algorithms for linearly constrained convex minimization and saddle-point problems: a unified approach
    Guoyong Gu
    Bingsheng He
    Xiaoming Yuan
    Computational Optimization and Applications, 2014, 59 : 135 - 161
  • [36] Customized proximal point algorithms for linearly constrained convex minimization and saddle-point problems: a unified approach
    Gu, Guoyong
    He, Bingsheng
    Yuan, Xiaoming
    COMPUTATIONAL OPTIMIZATION AND APPLICATIONS, 2014, 59 (1-2) : 135 - 161
  • [37] Online First-Order Framework for Robust Convex Optimization
    Ho-Nguyen, Nam
    Kilinc-Karzan, Fatma
    OPERATIONS RESEARCH, 2018, 66 (06) : 1670 - 1692
  • [38] Unified linear convergence of first-order primal-dual algorithms for saddle point problems
    Jiang, Fan
    Wu, Zhongming
    Cai, Xingju
    Zhang, Hongchao
    OPTIMIZATION LETTERS, 2022, 16 (06) : 1675 - 1700
  • [39] Unified linear convergence of first-order primal-dual algorithms for saddle point problems
    Fan Jiang
    Zhongming Wu
    Xingju Cai
    Hongchao Zhang
    Optimization Letters, 2022, 16 : 1675 - 1700
  • [40] Variance amplification of accelerated first-order algorithms for strongly convex quadratic optimization problems
    Mohammadi, Hesameddin
    Razaviyayn, Meisam
    Jovanovic, Mihailo R.
    2018 IEEE CONFERENCE ON DECISION AND CONTROL (CDC), 2018, : 5753 - 5758