Investigation of mutation operators for the bayesian optimization algorithm
1
Citation
16
Reference
10
Related Paper
Citation Trend
Abstract:
Although BOA is effective at finding solutions for optimization problems, small population sizes in a model can result in premature convergence to a sub-optimal solution. One way of avoiding premature convergence is to increase population diversity with a mutation operator. In our experiments, we compare several mutation operators for use with BOA. We examine in detail the probabilistic model utilizing (PMU) bit flipping mutation operator. We compare the effectiveness of the PMU operator with standard BOA, self-adaptive evolution and local search of substructural neighborhoods.Keywords:
Premature convergence
Operator (biology)
Adaptive mutation
In order to overcome the drawback of basic particle swarm optimization (PSO) such as being subject to being poor in performance of precision and falling into local optimization,a modified PSO is proposed by inducing adaptive mutation operator and clone operator in PSO.The selection operator can improve the fitness of the particle swarm to enhance the searching ability of arithmetic in local.The mutation operator can enlarge the searching scope to avoid premature convergence.The particle swarm will fly to the most optimization by adaptively adjusting the selection operator and mutation operator according to the change of the global extremum.The experiment results for typical functions show that the modified PSO can improve the performance of precision and avoid the premature convergence.
Premature convergence
Operator (biology)
Adaptive mutation
Local optimum
Cite
Citations (0)
We present the induced generalized probabilistic ordered weighted averaging (IGPOWA) operator. It is a new aggregation operator that unifies the probability and the OWA operator in the same formulation considering the degree of importance that each concept has in the formulation. Moreover, it also uses order-inducing variables in order to assess complex reordering processes. Furthermore, it also uses generalized means providing a more general framework that includes a wide range of particular cases such as the probabilistic aggregation, the maximum probabilistic aggregation, the minimum probabilistic aggregation, the arithmetic OWA and the arithmetic probabilistic aggregation. We further generalize this approach by using quasi-arithmetic means obtaining the quasi-arithmetic IPOWA (Quasi-IPOWA) operator. We also present an application of the new approach in a decision making problem.
Operator (biology)
Cite
Citations (1)
There are three difficult problems in the application of genetic algorithm, namely the parameter control, the premature convergence and the deception problem. Based on genetic algorithm with varying population size, a self adaptive genetic algorithm called natural genetic algorithm (nGA) is proposed. It introduces the population size threshold and the immigrant concepts, and adopts dynamically changing parameters in this paper. The design and structure of the nGA are discussed, and the performance of nGA is also analyzed.
Premature convergence
Cultural algorithm
Cite
Citations (1)
Operator (biology)
Group Decision Making
Cite
Citations (59)
In order to solve two difficult problems of premature convergence and slow searching speed of genetic algorithms in evolution neural network, a heuristic mutation operator is presented in this paper. Adaptive probability of mutation and heuristic mutation points selected is applied in it. When no evolution appears after many generations, the range of search will be extended by increasing probability of mutation, and a fine search will be started. The experiments of XOR problem demonstrate that the operator has fine ability of speedy convergence and maintains the diversity of the population automatically.
Premature convergence
Adaptive mutation
Operator (biology)
Cite
Citations (4)
Aiming at the phenomenon of premature convergence and later period oscillatory occurrences, an adaptive particle swarm optimization algorithm with the changes of the population diversity was proposed. In the algorithm, the adaptive exponent decreasing inertia weight and a dynamic adaptive changing threshold were proposed, the satisfied particle of threshold will be mutation by the average distance of particle. Adaptive adjustment of the threshold and the mutation can enhance the algorithm escape from local optima. The results show that the new algorithm of the global search capability has been improved, effectively avoid the premature convergence and later period oscillatory occurrences.
Premature convergence
Local optimum
Adaptive mutation
Exponent
Cite
Citations (8)
Considering the premature convergence problem of particle swarm optimization,a new adaptive particle swarm optimization is presented based on adaptive multiple mutation.The mutation probability for the current best particle is determined by two factors,including the variance of the population’s fitness and the current optimal solution.The ability of particle swarm optimization algorithm to break away from the local optinum is greatly improved by the mutation.A good performance of the algorithm is ensured in theory.The experimental results show that the new algorithm of global search capability not only is improved significantly,has an optimal convergence rate,but also can avoid the premature convergence problem effectively,and theory analysis show that it is feasible and availability.
Premature convergence
Adaptive mutation
Cite
Citations (0)
Considering the premature convergence problem of Particle Swarm Optimization(PSO),a new Adaptive Particle Swarm Optimization with Mutation(APSOwM) is presented based on the variance ratio of population’s fitness.During the running time,the inertia weight and the mutation probability are determined by two factors: the variance ratio of population’s fitness and the average distance of current population.The ability of APSOwM to break away from the local optimum and to find the global optimum is greatly improved by the adaptive mutation.Experimental results show that the new algorithm is with great advantage of convergence property over PSO,and also avoids the premature convergence problem effectively.
Premature convergence
Adaptive mutation
Cite
Citations (26)
An improved genetic algorithm is developed to solve the problems of the standard genetic algorithm easy to produce premature convergence and easy to fall into the local optimum,which has been successfully applied in east China city water supply optimization dispatch.The simulating results indicate solving optimal dispatch problems of large-scale water supply systems with the improved genetic algorithm can prevent premature and realize global-optimization effectively.
Premature convergence
Cite
Citations (0)