Analysis of an evolutionary algorithm with HyperMacromutation and stop at first constructive mutation heuristic for solving trap functions (original) (raw)

Evolutionary Algorithm Based on Automatically Designing of Genetic Operators

2013 Ninth International Conference on Computational Intelligence and Security, 2013

present there is a wide range of evolutionary algorithms available to researchers and practitioners. Despite the great diversity of these algorithms, virtually all of the algorithms share one feature: they have been manually designed. Can evolutionary algorithms be designed automatically by computer? In this paper, a novel evolutionary algorithm based on automatically designing of genetic operators is presented to address this problem. The resulting algorithm not only explores solutions in the problem space, but also automatically generates genetic operators in the operator space for each generation. In order to verify the performance of the proposed algorithm, comprehensive experiments on 23 well-known benchmark optimization problems are conducted, and the results show that the proposed algorithm can outperform standard Differential Evolution (DE) algorithm.

Enhanced Black Widow Algorithm for Numerical Functions Optimization

Revue d'Intelligence Artificielle, 2022

Black widow optimization algorithm is a recently evolutionary metaheuristic that imitates the unique mating behaviour of the black widow spiders in the real life. The trend of published papers utilizing the BWO algorithm is growing rapidly due to its efficiency in solving various engineering optimization problems. However, the BWO does not always perform as well as it should, and this is due to the random initialization of the spiders also the loss of good candidate solutions during the search. To remedy these problems, we propose in this paper a modified black widow optimization algorithm (MBWO) based on three ideas. First, an efficient initialization technique is adopted, which can guarantee starting the search with finest quality spiders and plays a significant role in determining an optimal or near-optimal solution. Second, the sexual cannibalism phase is modified to avoid the loss of high-quality solutions. Finally, an adaptive adjustment of crossover and mutation probabilities...

A New Genetic-Based Approach for Function Optimization

Advanced Materials Research, 2014

We present a novel genetic-based algorithm for optimizing n-D simple-bounded continuous functions. In this paper, we propose a new mutation operator, called rotational mutation. The proposed approach starts from the vertices of the polytope created by the simple bounds, as the initial population. Similar to the conventional genetic algorithm, we calculate the optimum point of each population based on its cost value using the elitism mechanism. Then, we create the new generations based on the proposed rotational mutation and the conventional crossover operators. We have evaluated the algorithm on the two well-known test problems. Experimental results showed that the proposed approach outperforms the conventional genetic algorithm, in terms of the number of generations.

An Efficient Genetic Algorithm for Numerical Function Optimization with Two New Crossover Operators

International Journal of Mathematical Sciences and Computing, 2018

Selection criteria, crossover and mutation are three main operators of genetic algorithm's performance. A lot of work has been done on these operators, but the crossover operator has a vital role in the operation of genetic algorithms. In literature, multiple crossover operators already exist with varying impact on the final results. In this article, we propose two new crossover operators for the genetic algorithms. One of them is based on the natural concept of crossover i.e. the upcoming offspring takes one bit from a parent and next from other parent and continuously takes bits till last one. The other proposed scheme is the extension of two-point crossover with the concept of multiplication rule. These operators are applied for eight benchmark problems in parallel with some traditional crossover operators. Empirical studies show a remarkable performance of the proposed crossover operators.

A Novel Hybrid Genetic Algorithm for Unconstrained and Constrained Function Optimization

Bio-Inspired Computing for Information Retrieval Applications, 2000

During the past decade, academic and industrial communities are highly interested in evolutionary techniques for solving optimization problems. Genetic Algorithm (GA) has proved its robustness in solving all most all types of optimization problems. To improve the performance of GA, several modifications have already been done within GA. Recently GA has been hybridized with many other nature-inspired algorithms. As such Bacterial Foraging Optimization (BFO) is popular bio inspired algorithm based on the foraging behavior of E. coli bacteria. Many researchers took active interest in hybridizing GA with BFO. Motivated by such popular hybridization of GA, an attempt has been made in this chapter to hybridize GA with BFO in a novel fashion. The Chemo-taxis step of BFO plays a major role in BFO. So an attempt has been made to hybridize Chemo-tactic step with GA cycle and the algorithm is named as Chemo-inspired Genetic Algorithm (CGA). It has been applied on benchmark functions and real life application problem to prove its efficacy.

Improved genetic algorithm inspired by biological evolution

2007

The process of mutation has been studied extensively in the field of biology and it has been shown that it is one of the major factors that aid the process of evolution. Inspired by this a novel genetic algorithm (GA) is presented here. Various mutation operators such as small mutation, gene mutation and chromosome mutation have been applied in this genetic algorithm. In order to facilitate the implementation of the above-mentioned mutation operators a modified way of representing the variables has been presented. It resembles the way genetic information is coded in living beings. Different mutation operators pose a challenge as regards the determination of the optimal rate of mutation. This problem is overcome by using adaptive mutation operators. The main purpose behind this approach was to improve the efficiency of GAs and to find widely distributed Pareto-optimal solutions. This algorithm was tested on some benchmark test functions and compared with other GAs. It was observed that the introduction of these mutations do improve the genetic algorithms in terms of convergence and the quality of the solutions.

New evolutionary optimization techniques and test functions for their evaluation

2014

The process of optimization is finding the best solution to a given problem, when the amount of available resources is often restricted. Despite the rapid development of computer science, most optimization problems can't be solved by evaluating all feasible solutions. For example, the class of NP-hard problems, such as the Traveling salesman problem (also known as TSP), might have an enormously large search space, which requires exponential computation time to be fully explored. To solve these kinds of problems, many heuristic algorithms have been developed. Heuristic algorithms can find approximate solutions, even when the search space is excessively huge. In this paper, we benchmarked twelve optimization techniques, to compare their efficiency in finding the global minima of different continuous mathematical test functions. Mathematical function optimization is very important, because most real world optimization problems can be modelled in this general framework. Numerous mat...

A New Hybrid of Evolutionary and Conventional Optimization Algorithms

Combination of the Evolutionary and Conventional algorithms has opened a new horizon to Optimization algorithms. In the previous hybrid algorithm, firstly an evolutionary algorithm has been applied and then the obtained result has been employed as an initial guess for a conventional algorithm. In this method the advantages of each algorithm cannot be realized while they are running at the same time. In addition, it is not determined in which iteration the evolutionary algorithm should stop and the conventional one should start. Certainly an improper iteration selection leads to an inefficient hybrid algorithm. To overcome above shortages, this paper proposes a novel hybrid algorithm that uses the abilities of evolutionary and conventional algorithm simultaneously. In each iteration of the proposed algorithm both evolutionary and conventional algorithms have been applied. Simulation results on some benchmark problems show that the proposed hybrid 816 A. Khosravi, A. Lari and J. Addeh algorithm has faster convergence and a more reliable solution than the conventional hybrid algorithm.