Dual Mutation Strategies and Dual Crossover Strategies for Differential Evolution
0
Citation
14
Reference
10
Related Paper
Abstract:
In this paper, there are two mutation strategies and two crossover strategies are involved for enhancing solution searching ability of Differential Evolution (DE). These strategies will be activated according to current solution searching status. The elitist mutation will guide particles toward to solution space around the elitist particles, and the random to real-rand mutation can prevent particles form fall into local optimum. Both elitist crossover and one-cut-point crossover can produce potential particles for deeply search the basin of solution space. In the experiments, 25 test functions of CEC 2005 are adopted for testing performance of proposed method and compare it with 4 DE variants. From the results, it can be observed that the proposed method exhibits better than related works for solving most test functions.Keywords:
Differential Evolution
Differential Evolution
Operator (biology)
Binomial (polynomial)
Cite
Citations (1)
Genetic algorithm (GA) is an artificial intelligence search method that uses the process of evolution and natural selection theory and is under the umbrella of evolutionary computing algorithm. It is an efficient tool for solving optimization problems. Integration among (GA) parameters is vital for successful (GA) search. Such parameters include mutation and crossover rates in addition to population that are important issues in (GA). However, each operator of GA has a special and different influence. The impact of these factors is influenced by their probabilities; it is difficult to predefine specific ratios for each parameter, particularly, mutation and crossover operators. This paper reviews various methods for choosing mutation and crossover ratios in GAs. Next, we define new deterministic control approaches for crossover and mutation rates, namely Dynamic Decreasing of high mutation ratio/dynamic increasing of low crossover ratio (DHM/ILC), and Dynamic Increasing of Low Mutation/Dynamic Decreasing of High Crossover (ILM/DHC). The dynamic nature of the proposed methods allows the ratios of both crossover and mutation operators to be changed linearly during the search progress, where (DHM/ILC) starts with 100% ratio for mutations, and 0% for crossovers. Both mutation and crossover ratios start to decrease and increase, respectively. By the end of the search process, the ratios will be 0% for mutations and 100% for crossovers. (ILM/DHC) worked the same but the other way around. The proposed approach was compared with two parameters tuning methods (predefined), namely fifty-fifty crossover/mutation ratios, and the most common approach that uses static ratios such as (0.03) mutation rates and (0.9) crossover rates. The experiments were conducted on ten Traveling Salesman Problems (TSP). The experiments showed the effectiveness of the proposed (DHM/ILC) when dealing with small population size, while the proposed (ILM/DHC) was found to be more effective when using large population size. In fact, both proposed dynamic methods outperformed the predefined methods compared in most cases tested.
Adaptive mutation
Cite
Citations (461)
In this paper, there are two mutation strategies and two crossover strategies are involved for enhancing solution searching ability of Differential Evolution (DE). These strategies will be activated according to current solution searching status. The elitist mutation will guide particles toward to solution space around the elitist particles, and the random to real-rand mutation can prevent particles form fall into local optimum. Both elitist crossover and one-cut-point crossover can produce potential particles for deeply search the basin of solution space. In the experiments, 25 test functions of CEC 2005 are adopted for testing performance of proposed method and compare it with 4 DE variants. From the results, it can be observed that the proposed method exhibits better than related works for solving most test functions.
Differential Evolution
Cite
Citations (0)
Differential Evolution
Perceptron
Backpropagation
Cite
Citations (7)
It is well known that a judicious choice of crossover and/or mutation rates is critical to the success of genetic algorithms. Most earlier researches focused on finding optimal crossover or mutation rates, which vary for different problems, and even for different stages of the genetic process in a problem. In this paper, a generic scheme for adapting the crossover and mutation probabilities is proposed. The crossover and mutation rates are adapted in response to the evaluation results of the respective offspring in the next generation. Experimental results show that the proposed scheme significantly improves the performance of genetic algorithms and outperforms previous work.
Cite
Citations (134)
Dual Population Differential Evolution algorithm based on Crossover and Mutation strategy(CMDPDE) is proposed to enhance global search ability of single population differential evolution.In CMDPDE,one population uses big scale factor and crossover factor,the other with small scale factor and crossover factor will execute crossover or mutation operations to search better individual after an evolution for each individual evolves one time per generation.At the same time evolution information will be exchanged between two populations after all individuals of two populations evolve ten times.Compared with single population differential evolution,CMDPDE increases diversity of solutions through dual population and crossover and mutation strategy,which makes CMDPDE search better solutions in a larger range.Experiment results on six benchmark functions show that CMDPDE has the better ability of finding optimal solution.
Differential Evolution
Benchmark (surveying)
Scale factor (cosmology)
Factor (programming language)
Cite
Citations (2)
Differential Evolution
Cite
Citations (63)
Differential Evolution
Cite
Citations (244)
In this paper, a Neural Networks optimizer based on Self-adaptive Differential Evolution is presented. This optimizer applies mutation and crossover operators in a new way, taking into account the structure of the network according to a per layer strategy. Moreover, a new crossover called interm is proposed, and a new self-adaptive version of DE called MAB-ShaDE is suggested to reduce the number of parameters. The framework has been tested on some well-known classification problems and a comparative study on the various combinations of self-adaptive methods, mutation, and crossover operators available in literature is performed. Experimental results show that DENN reaches good performances in terms of accuracy, better than or at least comparable with those obtained by backpropagation.
Differential Evolution
Backpropagation
Adaptive mutation
Cite
Citations (50)
Crossover is a very important operation in current differential evolution (DE) algorithms. The existing crossover strategies in DE show promising effects especially when the algorithms are applied to separable functions. However, the operation fails to work well when applied to the ill-conditioned and inseparable problems because the recombination of good genes is no longer promising for generating better individuals when the genes are highly correlated. Thus it is possible to use coordinate system rotation strategy which makes the variables be less correlated to improve the performance of the crossover operation. In this paper, we propose to use the principal component analysis (PCA) technique to rebuild a coordinate system. With this system, the correlations among variables are decreased for the crossover operation of DE. In every generation, the population and the mutated population are rotated into the new coordinate system to perform the crossover and then the newly generated population is rotated back to be evaluated. The PCA-based crossover is tested on the JADE algorithm. Experimental results show that the proposed method is quite promising on some ill-conditioned and inseparable functions.
Differential Evolution
Cite
Citations (4)