Topological Effects on the Performance of Island Model of Parallel Genetic Algorithm

The topological features of the communication network between computing nodes in Parallel Genetic Algorithms, under the framework of the island model, is discussed in the context of both the local rate of information exchange between nodes, and the global exchange rate that measures the level of information flow in the entire network. For optimal performance of parallel genetic algorithm for a set of benchmark functions, the connectivity of the network can be found, corresponding to a global information exchange rate between 40-70%. This range is obtained by statistical analysis on the search for solutions of four benchmark problems: the 0-1 knapsack, the Weierstrass's function, the Ackley's function, and the Modified Shekel's foxholes function. Our method is based on the cutting of links of a fully connected network to gradually decrease the connectivity, and compare the performance of the genetic algorithm on each network. Suggestions for the protocol in applying this general guideline in the design of a good communication network for parallel genetic algorithms are made, where the islands are connected with 40% of links of a fully connected network before fine tuning the parameters of the island model to enhance performance in a specific problem.

[1]  D. Ackley.  A connectionist machine for genetic hillclimbing , 1987 .

[2]  David E. Goldberg,et al.  Genetic Algorithms in Search Optimization and Machine Learning , 1988 .

[3]  Jose Miguel Puerta,et al.  Initial approaches to the application of islands-based parallel EDAs in continuous domains , 2005, 2005 International Conference on Parallel Processing Workshops (ICPPW'05).

[4]  K. Szeto,et al.  Locus Oriented Adaptive Genetic Algorithm: Application to the Zero/One Knapsack Problem , 2004 .

[5]  Luca Maria Gambardella,et al.  Results of the first international contest on evolutionary optimisation (1st ICEO) , 1996, Proceedings of IEEE International Conference on Evolutionary Computation.

[6]  Matthias Scheutz,et al.  Adaptive algorithms for the dynamic distribution and parallel execution of agent-based models , 2006, J. Parallel Distributed Comput..

[7]  Kwok Yip Szeto,et al.  Adaptive Genetic Algorithm with Mutation and Crossover Matrices , 2007, IJCAI.

[8]  P. Deuflhard,et al.  Large Scale Scientific Computing , 1987 .

[9]  Jian Zhang,et al.  Adaptive Genetic Algorithm and Quasi-parallel Genetic Algorithm: Application to Knapsack Problem , 2005, LSSC.

[10]  Kwok Yip Szeto,et al.  Sales Potential Optimization on Directed Social Networks: A Quasi-Parallel Genetic Algorithm Approach , 2012, EvoApplications.

[11]  Kwok Yip Szeto,et al.  Importance of information exchange in quasi-parallel genetic algorithms , 2011, GECCO '11.

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

[13]  Andre B. Bondi,et al.  Characteristics of scalability and their impact on performance , 2000, WOSP '00.

[14]  Kwok Yip Szeto,et al.  Self-adaptive Mutation Only Genetic Algorithm: An Application on the Optimization of Airport Capacity Utilization , 2008, IDEAL.

[15]  Kwok Yip Szeto,et al.  Rules extraction in short memory time series using genetic algorithms , 2001 .

[16]  El-Ghazali Talbi,et al.  Grid computing for parallel bioinspired algorithms , 2006, J. Parallel Distributed Comput..

[17]  Edmund K. Burke,et al.  The Speciating Island Model: An alternative parallel evolutionary algorithm , 2006, J. Parallel Distributed Comput..

[18]  John H. Holland,et al.  Adaptation in Natural and Artificial Systems: An Introductory Analysis with Applications to Biology, Control, and Artificial Intelligence , 1992 .

[19]  Erick Cantú-Paz,et al.  Efficient and Accurate Parallel Genetic Algorithms , 2000, Genetic Algorithms and Evolutionary Computation.