Memetic clonal selection algorithm with EDA vaccination for unconstrained binary quadratic programming problems

  • Authors:

  • Yiqiao Cai;Jiahai Wang;Jian Yin;Yalan Zhou

  • Affiliations:

  • Department of Computer Science, Sun Yat-sen University, No. 132, Waihuan East Road, Guangzhou Higher Education Mega Center 510006, PR China;Department of Computer Science, Sun Yat-sen University, No. 132, Waihuan East Road, Guangzhou Higher Education Mega Center 510006, PR China;Department of Computer Science, Sun Yat-sen University, No. 132, Waihuan East Road, Guangzhou Higher Education Mega Center 510006, PR China;Information Science School, Guangdong University of Business Studies, No. 21, Chisha Road, Guangzhou 510320, PR China

  • Venue:
  • Expert Systems with Applications: An International Journal
  • Year:
  • 2011

Quantified Score

Hi-index 12.05

Visualization

Abstract

This paper presents a memetic clonal selection algorithm (MCSA) with estimation of distribution algorithm (EDA) vaccination, named MCSA-EDA, for the unconstrained binary quadratic programming problem (UBQP). In order to improve the performance of the conventional clonal selection algorithm (CSA), three components are adopted in MCSA-EDA. First, to compensate for the absence of recombination among different antibodies, an EDA vaccination is designed and incorporated into CSA. Second, to keep the diversity of the population, a fitness uniform selection scheme (FUSS) is adopted as a selection operator. Third, to enhance the exploitation ability of CSA, an adaptive tabu search (TS) with feedback mechanism is introduced. Thus, MCSA-EDA can overcome the deficiencies of CSA and further search better solutions. MCSA-EDA is tested on a series of UBQP with size up to 7000 variables. Simulation results show that MCSA-EDA is effective for improving the performance of the conventional CSA and is better than or at least competitive with other existing metaheuristic algorithms.