The Exploration/Exploitation Tradeoff in Dynamic Cellular Genetic Algorithms

Abstract

Abstract-This paper studies static and dynamic decentralized versions of the search model known as cellular genetic algorithm (cGA), in which individuals are located in a specific topology and interact only with their neighbors. Making changes in the shape of such topology or in the neighborhood may give birth to a high number of algorithmic variants. We perform these changes in a methodological way by tuning the concept of ratio. Since the relationship (ratio) between the topology and the neighborhood shape defines the search selection pressure, we propose to analyze in depth the influence of this ratio on the exploration/exploitation tradeoff. As we will see, it is difficult to decide which ratio is best suited for a given problem. Therefore, we introduce a preprogrammed change of this ratio during the evolution as a possible additional improvement that removes the need of specifying a single ratio. A later refinement will lead us to the first adaptive dynamic kind of cellular models to our knowledge. We conclude that these dynamic cGAs have the most desirable behavior among all the evaluated ones in terms of efficiency and accuracy; we validate our results on a set of seven different problems of considerable complexity in order to better sustain our conclusions.Index Terms-Cellular genetic algorithm (cGA), evolutionary algorithm (EA), dynamic adaptation, neighborhood-to-population ratio.

Dynamic and Partially Connected Ring Topologies for Evolutionary Algorithms with Structured Populations

“…Depending on its parameters, trap functions may be deceptive or not. The trap functions in these experiments are defined by: (1) where u( ) is the unitation function and is the problem size (and also the fitness of the global optimum). With these definitions, order-traps are in the region between deceptive and non-deceptive, while order-are non-deceptive and order-4 are fully deceptive.…”

“…Parameterization was done after [1]: population size was set to ; the recombination operator is the double point crossover with ; mutation is bit-flip with , where is the chromosome length. Only one offspring is placed in the temporal population (randomly chosen from the set of two children).…”

Oversized Populations and Cooperative Selection: Dealing with Massive Resources in Parallel Infrastructures

“…In addition to canonical approaches such as ranking, roulette wheel or tournament selection [3], other selection schemes have been designed to trade off exploration and exploitation [1] or to be able to self-adapt the selection pressure on-line [2], just to mention a few. Cooperation has been also considered in the design of co-evolutionary EAs [9] in which sub-populations represent partial solutions to a problem and have to collaborate in order to build up complete solutions.…”

The Effects of Heterogeneity on Asynchronous Panmictic Genetic Search

“…Island [3,5] approaches evaluate local populations for a certain number of iterations, then exchange the best members with other islands. Cellular algorithms [2,9] evaluate individual parameter sets, then update these individual sets based on the fitness of their neighbors. Hybrid approaches [14,18] have also been examined.…”

“…In the first phase of the algorithm (while the population size is less than the maximum population size) the server is being initialized and a random popu- .params -m1.params result [1].params = diff -m1.params result [2].params = diff + m2.params return result lation is generated. When a request work message is processed, a random parameter set is generated, and when a report work message is processed, the population is updated with the parameters and the fitness of that evaluation.…”