Random drift particle swarm optimization algorithm: convergence analysis and parameter selection

J. Sun, X. Wu, V. Palade, W. Fang, Y. Shi

    Research output: Contribution to journalArticle

    23 Citations (Scopus)
    47 Downloads (Pure)

    Abstract

    The random drift particle swarm optimization (RDPSO) algorithm is a PSO variant inspired by the free electron model in metal conductors placed in an external electric field. Based on the preliminary work on the RDPSO algorithm, this paper makes systematical analyses and empirical studies of the algorithm. Firstly, the motivation of the RDPSO algorithm is presented and the design of the particle’s velocity equation is described in detail. Secondly, a comprehensive analysis of the algorithm is made in order to gain a deep insight into how the RDPSO algorithm works. It involves a theoretical analysis and the simulation of the stochastic dynamical behavior of a single particle in the RDPSO algorithm. The search behavior of the algorithm itself is also investigated in detail, by analyzing the interaction among the particles. Then, some variants of the RDPSO algorithm are presented by incorporating different random velocity components with different neighborhood topologies. Finally, empirical studies of the RDPSO algorithm are performed by using a set of benchmark functions from the CEC2005 benchmark suite. Based on the theoretical analysis of the particle’s behavior, two methods of controlling the algorithmic parameters are employed, followed by an experimental analysis on how to select the parameter values, in order to obtain a satisfactory overall performance of the RDPSO algorithm and its variants in real-world applications. A further performance comparison between the RDPSO algorithm and other variants of PSO is made to prove the effectiveness of the RDPSO.
    Original languageEnglish
    Pages (from-to)345-376
    JournalMachine Learning
    Volume101
    Issue number1-3
    Early online date15 Aug 2015
    DOIs
    Publication statusPublished - Oct 2015

    Fingerprint

    Particle swarm optimization (PSO)
    Set theory
    Electric fields
    Topology
    Electrons

    Bibliographical note

    The final publication is available at Springer via http://dx.doi.org/10.1007/s10994-015-5522-z

    Keywords

    • Evolutionary computation
    • Optimization
    • Particle swarm optimization
    • Random motion

    Cite this

    Random drift particle swarm optimization algorithm: convergence analysis and parameter selection. / Sun, J.; Wu, X.; Palade, V.; Fang, W.; Shi, Y.

    In: Machine Learning, Vol. 101, No. 1-3, 10.2015, p. 345-376.

    Research output: Contribution to journalArticle

    Sun, J. ; Wu, X. ; Palade, V. ; Fang, W. ; Shi, Y. / Random drift particle swarm optimization algorithm: convergence analysis and parameter selection. In: Machine Learning. 2015 ; Vol. 101, No. 1-3. pp. 345-376.
    @article{0c65293d7660435dacfcea2f30dc595a,
    title = "Random drift particle swarm optimization algorithm: convergence analysis and parameter selection",
    abstract = "The random drift particle swarm optimization (RDPSO) algorithm is a PSO variant inspired by the free electron model in metal conductors placed in an external electric field. Based on the preliminary work on the RDPSO algorithm, this paper makes systematical analyses and empirical studies of the algorithm. Firstly, the motivation of the RDPSO algorithm is presented and the design of the particle’s velocity equation is described in detail. Secondly, a comprehensive analysis of the algorithm is made in order to gain a deep insight into how the RDPSO algorithm works. It involves a theoretical analysis and the simulation of the stochastic dynamical behavior of a single particle in the RDPSO algorithm. The search behavior of the algorithm itself is also investigated in detail, by analyzing the interaction among the particles. Then, some variants of the RDPSO algorithm are presented by incorporating different random velocity components with different neighborhood topologies. Finally, empirical studies of the RDPSO algorithm are performed by using a set of benchmark functions from the CEC2005 benchmark suite. Based on the theoretical analysis of the particle’s behavior, two methods of controlling the algorithmic parameters are employed, followed by an experimental analysis on how to select the parameter values, in order to obtain a satisfactory overall performance of the RDPSO algorithm and its variants in real-world applications. A further performance comparison between the RDPSO algorithm and other variants of PSO is made to prove the effectiveness of the RDPSO.",
    keywords = "Evolutionary computation, Optimization, Particle swarm optimization, Random motion",
    author = "J. Sun and X. Wu and V. Palade and W. Fang and Y. Shi",
    note = "The final publication is available at Springer via http://dx.doi.org/10.1007/s10994-015-5522-z",
    year = "2015",
    month = "10",
    doi = "10.1007/s10994-015-5522-z",
    language = "English",
    volume = "101",
    pages = "345--376",
    journal = "Machine Learning",
    issn = "0885-6125",
    publisher = "Springer Verlag",
    number = "1-3",

    }

    TY - JOUR

    T1 - Random drift particle swarm optimization algorithm: convergence analysis and parameter selection

    AU - Sun, J.

    AU - Wu, X.

    AU - Palade, V.

    AU - Fang, W.

    AU - Shi, Y.

    N1 - The final publication is available at Springer via http://dx.doi.org/10.1007/s10994-015-5522-z

    PY - 2015/10

    Y1 - 2015/10

    N2 - The random drift particle swarm optimization (RDPSO) algorithm is a PSO variant inspired by the free electron model in metal conductors placed in an external electric field. Based on the preliminary work on the RDPSO algorithm, this paper makes systematical analyses and empirical studies of the algorithm. Firstly, the motivation of the RDPSO algorithm is presented and the design of the particle’s velocity equation is described in detail. Secondly, a comprehensive analysis of the algorithm is made in order to gain a deep insight into how the RDPSO algorithm works. It involves a theoretical analysis and the simulation of the stochastic dynamical behavior of a single particle in the RDPSO algorithm. The search behavior of the algorithm itself is also investigated in detail, by analyzing the interaction among the particles. Then, some variants of the RDPSO algorithm are presented by incorporating different random velocity components with different neighborhood topologies. Finally, empirical studies of the RDPSO algorithm are performed by using a set of benchmark functions from the CEC2005 benchmark suite. Based on the theoretical analysis of the particle’s behavior, two methods of controlling the algorithmic parameters are employed, followed by an experimental analysis on how to select the parameter values, in order to obtain a satisfactory overall performance of the RDPSO algorithm and its variants in real-world applications. A further performance comparison between the RDPSO algorithm and other variants of PSO is made to prove the effectiveness of the RDPSO.

    AB - The random drift particle swarm optimization (RDPSO) algorithm is a PSO variant inspired by the free electron model in metal conductors placed in an external electric field. Based on the preliminary work on the RDPSO algorithm, this paper makes systematical analyses and empirical studies of the algorithm. Firstly, the motivation of the RDPSO algorithm is presented and the design of the particle’s velocity equation is described in detail. Secondly, a comprehensive analysis of the algorithm is made in order to gain a deep insight into how the RDPSO algorithm works. It involves a theoretical analysis and the simulation of the stochastic dynamical behavior of a single particle in the RDPSO algorithm. The search behavior of the algorithm itself is also investigated in detail, by analyzing the interaction among the particles. Then, some variants of the RDPSO algorithm are presented by incorporating different random velocity components with different neighborhood topologies. Finally, empirical studies of the RDPSO algorithm are performed by using a set of benchmark functions from the CEC2005 benchmark suite. Based on the theoretical analysis of the particle’s behavior, two methods of controlling the algorithmic parameters are employed, followed by an experimental analysis on how to select the parameter values, in order to obtain a satisfactory overall performance of the RDPSO algorithm and its variants in real-world applications. A further performance comparison between the RDPSO algorithm and other variants of PSO is made to prove the effectiveness of the RDPSO.

    KW - Evolutionary computation

    KW - Optimization

    KW - Particle swarm optimization

    KW - Random motion

    U2 - 10.1007/s10994-015-5522-z

    DO - 10.1007/s10994-015-5522-z

    M3 - Article

    VL - 101

    SP - 345

    EP - 376

    JO - Machine Learning

    JF - Machine Learning

    SN - 0885-6125

    IS - 1-3

    ER -