Causes of the Imbalance Between Exploration and Exploitation in Evolutionary Computation

Evolutionary algorithms have been used in more and more research fields. However, it is very usual that an optimal of nontrivial problems cannot be found by an evolutionary algorithm. In fact, only if the balance between exploration and exploitation is achieved in runs, good solutions can be obtained. In this paper, we observe the changing trend of genotype diversity in runs, which cannot obtain the optimal, of different EAs. Then, we illustrate the main cause of the imbalance between exploration and exploitation in different situations.

[1]  Kenli Li,et al.  Hybrid immune algorithm based on greedy algorithm and delete-cross operator for solving TSP , 2014, Soft Computing.

[2]  Enrique Alba,et al.  Parallelism and evolutionary algorithms , 2002, IEEE Trans. Evol. Comput..

[3]  Marjan Mernik,et al.  Exploration and exploitation in evolutionary algorithms: A survey , 2013, CSUR.

[4]  Shengxiang Yang,et al.  An Ant Colony Optimization Based Memetic Algorithm for the Dynamic Travelling Salesman Problem , 2015, GECCO.

[5]  Jürgen Branke,et al.  Balancing Population- and Individual-Level Adaptation in Changing Environments , 2009, Adapt. Behav..

[6]  Gerhard Reinelt,et al.  TSPLIB - A Traveling Salesman Problem Library , 1991, INFORMS J. Comput..

[7]  Long Li,et al.  Differential evolution based on covariance matrix learning and bimodal distribution parameter setting , 2014, Appl. Soft Comput..

[8]  Gilbert Syswerda,et al.  Uniform Crossover in Genetic Algorithms , 1989, ICGA.

[9]  Shengxiang Yang,et al.  Ant Colony Optimization With Local Search for Dynamic Traveling Salesman Problems , 2017, IEEE Transactions on Cybernetics.

[10]  A. E. Eiben,et al.  On Evolutionary Exploration and Exploitation , 1998, Fundam. Informaticae.

[11]  Jing J. Liang,et al.  Problem Definitions and Evaluation Criteria for the CEC 2005 Special Session on Real-Parameter Optimization , 2005 .

[12]  Zbigniew Michalewicz,et al.  Inver-over Operator for the TSP , 1998, PPSN.

[13]  Graham Kendall,et al.  Diversity in genetic programming: an analysis of measures and correlation with fitness , 2004, IEEE Transactions on Evolutionary Computation.

[14]  Anthony Brabazon,et al.  Towards an understanding of locality in genetic programming , 2010, GECCO '10.

[15]  Dario Izzo,et al.  Evolving Solutions to TSP Variants for Active Space Debris Removal , 2015, GECCO.

[16]  Pei-Chann Chang,et al.  Dynamic diversity control in genetic algorithm for mining unsearched solution space in TSP problems , 2010, Expert Syst. Appl..