Nesterov Acceleration for Equality-Constrained Convex Optimization via Continuously Differentiable Penalty Functions

被引:8
|
作者
Srivastava, Priyank [1 ]
Cortes, Jorge [1 ]
机构
[1] Univ Calif San Diego, Dept Mech & Aerosp Engn, San Diego, CA 92122 USA
来源
IEEE CONTROL SYSTEMS LETTERS | 2021年 / 5卷 / 02期
关键词
Convex functions; Acceleration; Convergence; Linear programming; Gradient methods; Newton method; Constrained optimization; accelerated flows; smooth exact penalty functions; convex functions; SADDLE-POINT PROBLEMS; CONVERGENCE;
D O I
10.1109/LCSYS.2020.3002687
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
We propose a framework to use Nesterov's accelerated method for constrained convex optimization problems. Our approach consists of first reformulating the original problem as an unconstrained optimization problem using a continuously differentiable exact penalty function. This reformulation is based on replacing the Lagrange multipliers in the augmented Lagrangian of the original problem by Lagrange multiplier functions. The expressions of these Lagrange multiplier functions, which depend upon the gradients of the objective function and the constraints, can make the unconstrained penalty function non-convex in general even if the original problem is convex. We establish sufficient conditions on the objective function and the constraints of the original problem under which the unconstrained penalty function is convex. This enables us to use Nesterov's accelerated gradient method for unconstrained convex optimization and achieve a guaranteed rate of convergence which is better than the state-of-the-art first-order algorithms for constrained convex optimization. Simulations illustrate our results.
引用
收藏
页码:415 / 420
页数:6
相关论文
共 50 条
  • [1] ON DUAL DIFFERENTIABLE EXACT PENALTY-FUNCTIONS FOR EQUALITY CONSTRAINED OPTIMIZATION
    FUKUSHIMA, M
    YAMAKAWA, E
    MINE, H
    JOURNAL OF THE OPERATIONS RESEARCH SOCIETY OF JAPAN, 1985, 28 (04) : 302 - 317
  • [2] ON THE ROLE OF CONTINUOUSLY DIFFERENTIABLE EXACT PENALTY-FUNCTIONS IN CONSTRAINED GLOBAL OPTIMIZATION
    LUCIDI, S
    JOURNAL OF GLOBAL OPTIMIZATION, 1994, 5 (01) : 49 - 68
  • [3] IMPLEMENTING A SMOOTH EXACT PENALTY FUNCTION FOR EQUALITY-CONSTRAINED NONLINEAR OPTIMIZATION
    Estrin, Ron
    Friedlander, Michael P.
    Orban, Dominique
    Saunders, Michael A.
    SIAM JOURNAL ON SCIENTIFIC COMPUTING, 2020, 42 (03): : A1809 - A1835
  • [4] Equality-constrained minimization of polynomial functions
    ShuiJing Xiao
    GuangXing Zeng
    Science China Mathematics, 2015, 58 : 1 - 24
  • [5] Equality-constrained minimization of polynomial functions
    XIAO ShuiJing
    ZENG GuangXing
    ScienceChina(Mathematics), 2015, 58 (10) : 2181 - 2204
  • [6] Equality-constrained minimization of polynomial functions
    Xiao ShuiJing
    Zeng GuangXing
    SCIENCE CHINA-MATHEMATICS, 2015, 58 (10) : 2181 - 2204
  • [7] CONSTRAINED OPTIMIZATION VIA PENALTY FUNCTIONS
    LOOTSMA, FA
    PHILIPS RESEARCH REPORTS, 1968, 23 (05): : 408 - &
  • [8] Quasi-Newton acceleration for equality-constrained minimization
    Ferreira-Mendonca, L.
    Lopes, V. L. R.
    Martinez, J. M.
    COMPUTATIONAL OPTIMIZATION AND APPLICATIONS, 2008, 40 (03) : 373 - 388
  • [9] Quasi-Newton acceleration for equality-constrained minimization
    L. Ferreira-Mendonça
    V. L. R. Lopes
    J. M. Martínez
    Computational Optimization and Applications, 2008, 40 : 373 - 388
  • [10] Sequential equality-constrained optimization for nonlinear programming
    Birgin, E. G.
    Bueno, L. F.
    Martinez, J. M.
    COMPUTATIONAL OPTIMIZATION AND APPLICATIONS, 2016, 65 (03) : 699 - 721