Parallel two-phase methods for global optimization on GPU

被引:8
|
作者
Ferreiro, Ana M. [1 ,2 ]
Garcia-Rodriguez, Jose Antonio [1 ,2 ]
Vazquez, Carlos [1 ,2 ]
Costa e Silva, E. [3 ]
Correia, A. [3 ]
机构
[1] Fac Informat, Dept Math, CITIC, Campus Elvina S-N, La Coruna 15071, Spain
[2] ITMATI, La Coruna, Spain
[3] Porto Polytech, CIICESI ESTGF, Porto, Portugal
关键词
Global optimization; Basin Hopping; Conjugate gradient method; Parallelization; GPUs; MEAD SIMPLEX-METHOD; CONVERGENCE PROPERTIES; GENETIC ALGORITHM; PATTERN SEARCH; MINIMIZATION;
D O I
10.1016/j.matcom.2018.06.005
中图分类号
TP39 [计算机的应用];
学科分类号
081203 ; 0835 ;
摘要
Developing general global optimization algorithms is a difficult task, specially for functions with a huge number of local minima in high dimensions. Stochastic metaheuristic algorithms can provide the only alternative for the solution of such problems since they are aimed at guaranteeing global optimality. However, the main drawback of these algorithms is that they require a large number of function evaluations in order to skip/discard local optima, thus exhibiting a low convergence order and, as a result, a high computational cost. Furthermore, the situation can become even worse with the increase of dimension. Usually the number of local minima highly increases, as well as the computational cost of the function evaluation, thus increasing the difficulty for covering the whole search space. On the other hand, deterministic local optimization methods exhibit faster convergence rates, requiring a lower number of functions evaluations and therefore involving a lower computational cost, although they can get stuck into local minima. A way to obtain faster global optimization algorithms is to mix local and global methods in order to benefit from higher convergence rates of local ones, while retaining the global approximation properties. Another way to speedup global optimization algorithms comes from the use of efficient parallel hardware architectures. Nowadays, a good alternative is to take advantage of graphics processing units (GPUs), which are massively parallel processors and have become quite accessible cheap alternative for high performance computing. In this work a parallel implementation on GPUs of some hybrid two-phase optimization methods, that combine the metaheuristic Simulated Annealing algorithm for finding a global minimum, with different local optimization methods, namely a conjugate gradient algorithm and a version of Nelder-Mead method, is presented. The performance of parallelized versions of the above hybrid methods are analyzed for a set of well known test problems. Results show that GPUs represent an efficient alternative for the parallel implementation of two-phase global optimization methods. (C) 2018 International Association for Mathematics and Computers in Simulation (IMACS). Published by Elsevier B.V. All rights reserved.
引用
收藏
页码:67 / 90
页数:24
相关论文
共 50 条
  • [41] Optimization of a separator assisted two-phase thermosyphon loop by using entropy generation analysis
    Zhu, Lin
    Yu, Jianlin
    INTERNATIONAL JOURNAL OF REFRIGERATION-REVUE INTERNATIONALE DU FROID, 2019, 98 : 15 - 24
  • [42] Two-phase hybrid Particle Swarm Optimization Applied to the Double Row Layout Problem
    Cravo, Gildasio Lecchi
    Bissoli, Dayan de Castro
    Sales Amaral, Andre Renato
    INTELIGENCIA ARTIFICIAL-IBEROAMERICAL JOURNAL OF ARTIFICIAL INTELLIGENCE, 2021, 24 (67): : 51 - 70
  • [43] Steklov regularization and trajectory methods for univariate global optimization
    Arikan, Orhan
    Burachik, Regina S.
    Kaya, C. Yalcin
    JOURNAL OF GLOBAL OPTIMIZATION, 2020, 76 (01) : 91 - 120
  • [44] A Two-Phase PCBA Optimization With ILP Model and Heuristic for a Beam Head Placement Machine
    Lu, Guangyu
    Li, Zhengkai
    Sun, Hao
    Yu, Xinghu
    Qin, Jiahu
    Qiu, Jianbin
    Gao, Huijun
    IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS, 2024, 20 (12) : 13612 - 13621
  • [45] Global descent methods for unconstrained global optimization
    Wu, Z. Y.
    Li, D.
    Zhang, L. S.
    JOURNAL OF GLOBAL OPTIMIZATION, 2011, 50 (03) : 379 - 396
  • [46] Parallel Versions of the Modified Coordinate and Gradient Descent Methods and Their Application to a Class of Global Optimization Problems
    S. K. Zavriev
    Yu. N. Perunova
    Computational Mathematics and Modeling, 2003, 14 (2) : 108 - 122
  • [47] Solving nonlinear constrained optimization problems: An immune evolutionary based two-phase approach
    Hsieh, Yi-Chih
    Lee, Yung-Cheng
    You, Peng-Sheng
    APPLIED MATHEMATICAL MODELLING, 2015, 39 (19) : 5759 - 5768
  • [48] A Two-Phase Optimization System Applied in Optical Design of LED Light Guide Plate
    Chen, Chen-Tai
    Chen, Bai-Rui
    INTERNATIONAL CONFERENCE ON ENGINEERING AND BUSINESS MANAGEMENT (EBM2011), VOLS 1-6, 2011, : 3219 - +
  • [49] Two Modified PRP Conjugate Gradient Methods and Their Global Convergence for Unconstrained Optimization
    Sun Zhongbo
    Cao Xue
    Guo Yingying
    Ge Yuncheng
    Sun Yue
    2017 29TH CHINESE CONTROL AND DECISION CONFERENCE (CCDC), 2017, : 7896 - 790
  • [50] Two modified spectral conjugate gradient methods and their global convergence for unconstrained optimization
    Sun, Zhongbo
    Li, Hongyang
    Wang, Jing
    Tian, Yantao
    INTERNATIONAL JOURNAL OF COMPUTER MATHEMATICS, 2018, 95 (10) : 2082 - 2099