Hyper-Heuristic Approach for Tuning Parameter Adaptation in Differential Evolution

被引:2
|
作者
Stanovov, Vladimir [1 ,2 ]
Kazakovtsev, Lev [1 ,2 ]
Semenkin, Eugene [1 ,2 ]
机构
[1] Siberian Fed Univ, Lab Hybrid Methods Modelling & Optimizat Complex S, Krasnoyarsk 660074, Russia
[2] Reshetnev Siberian State Univ Sci & Technol, Inst Informat & Telecommun, Krasnoyarsk 660037, Russia
关键词
numerical optimization; differential evolution; parameter adaptation; hyper-heuristic; GLOBAL OPTIMIZATION; ALGORITHM;
D O I
10.3390/axioms13010059
中图分类号
O29 [应用数学];
学科分类号
070104 ;
摘要
Differential evolution (DE) is one of the most promising black-box numerical optimization methods. However, DE algorithms suffer from the problem of control parameter settings. Various adaptation methods have been proposed, with success history-based adaptation being the most popular. However, hand-crafted designs are known to suffer from human perception bias. In this study, our aim is to design automatically a parameter adaptation method for DE with the use of the hyper-heuristic approach. In particular, we consider the adaptation of scaling factor F, which is the most sensitive parameter of DE algorithms. In order to propose a flexible approach, a Taylor series expansion is used to represent the dependence between the success rate of the algorithm during its run and the scaling factor value. Moreover, two Taylor series are used for the mean of the random distribution for sampling F and its standard deviation. Unlike most studies, the Student's t distribution is applied, and the number of degrees of freedom is also tuned. As a tuning method, another DE algorithm is used. The experiments performed on a recently proposed L-NTADE algorithm and two benchmark sets, CEC 2017 and CEC 2022, show that there is a relatively simple adaptation technique with the scaling factor changing between 0.4 and 0.6, which enables us to achieve high performance in most scenarios. It is shown that the automatically designed heuristic can be efficiently approximated by two simple equations, without a loss of efficiency.
引用
收藏
页数:19
相关论文
共 50 条
  • [41] Hyper-heuristic Decision Tree Induction
    Vella, Alan
    Corne, David
    Murphy, Chris
    2009 WORLD CONGRESS ON NATURE & BIOLOGICALLY INSPIRED COMPUTING (NABIC 2009), 2009, : 408 - +
  • [42] An investigation and extension of a hyper-heuristic framework
    Rattadilok, Prapa
    Informatica (Ljubljana), 2010, 34 (04) : 523 - 534
  • [43] A hyper-heuristic for descriptive rule induction
    Faculty of Information Technology, Hanoi University of Education, Hanoi, Viet Nam
    不详
    Int. J. Data Warehouse. Min., 2007, 1 (54-66):
  • [44] Searching the Hyper-heuristic Design Space
    Swan, Jerry
    Woodward, John
    Ozcan, Ender
    Kendall, Graham
    Burke, Edmund
    COGNITIVE COMPUTATION, 2014, 6 (01) : 66 - 73
  • [45] Evolutionary Multilabel Hyper-Heuristic Design
    Rosales-Perez, Alejandro
    Gutierrez-Rodriguez, Andres E.
    Ortiz-Bayliss, Jose C.
    Terashima-Marin, Hugo
    Coello Coello, Carlos A.
    2017 IEEE CONGRESS ON EVOLUTIONARY COMPUTATION (CEC), 2017, : 2622 - 2629
  • [46] An analysis of heuristic subsequences for offline hyper-heuristic learning
    Yates, W. B.
    Keedwell, E. C.
    JOURNAL OF HEURISTICS, 2019, 25 (03) : 399 - 430
  • [47] Analysing Heuristic Subsequences for Offline Hyper-heuristic Learning
    Yates, William B.
    Keedwell, Edward C.
    PROCEEDINGS OF THE 2019 GENETIC AND EVOLUTIONARY COMPUTATION CONFERENCE COMPANION (GECCCO'19 COMPANION), 2019, : 37 - 38
  • [48] A cooperative hyper-heuristic search framework
    Djamila Ouelhadj
    Sanja Petrovic
    Journal of Heuristics, 2010, 16 : 835 - 857
  • [49] An investigation of hyper-heuristic search spaces
    Rodriguez, Jose Antonio Vazquez
    Petrovic, Sanja
    Salhi, Abdellah
    2007 IEEE CONGRESS ON EVOLUTIONARY COMPUTATION, VOLS 1-10, PROCEEDINGS, 2007, : 3776 - +
  • [50] An Investigation and Extension of a Hyper-heuristic Framework
    Rattadilok, Prapa
    INFORMATICA-JOURNAL OF COMPUTING AND INFORMATICS, 2010, 34 (04): : 523 - 534