Dynamic Niche Clustering: A Fuzzy Variable Radius Niching Technique for Multimodal Optimisation in GAs
نویسندگان
چکیده
This paper describes the recent developments and improvements made to the variable radius niching technique called Dynamic Niche Clustering (DNC) originally described in [4] and [5]. DNC employs a separate population of overlapping fuzzy niches with independent radii which operate in the decoded parameter space and are maintained alongside the normal GA population. We describe a speedup process that can be applied to the initial generation, which greatly reduces the complexity of the initial stages. A split operator is also introduced that is designed to counteract the excessive growth of niches, and it is shown that this improves the overall robustness of the technique. Finally, the effect of local elitism is documented and compared to the performance of the basic DNC technique on a selection of 2D test functions. The paper is concluded with a view to future work to be undertaken on the technique.
منابع مشابه
A Variable Radius Niching Technique for Speciation in Genetic Algorithms
In this paper, a continuation of a variable radius niche technique called Dynamic Niche Clustering developed by (Gan & Warwick, 1999) is presented. The technique employs a separate dynamic population of overlapping niches that coexists alongside the normal population. An empirical analysis of the updated methodology on a large group of standard optimisation test-bed functions is also given. The...
متن کاملA Genetic Algorithm with Sharing Scheme using Fuzzy Adaptive Clustering in Multimodal Function Optimization
Genetic Algorithms (GAs) are systems based upon principles from biological genetics that have been used in function optimization. However, traditional GAs have shown to be inadequate in some cases, specially multimodal functions. Niching Methods allow genetic algorithms to maintain a population of diverse individuals. GAs that incorporate these methods are capable of locating multiple, optimal ...
متن کاملModelling Niches of Arbitrary Shape in Genetic Algorithms using Niche Linkage in the Dynamic Niche Clustering Framework
We present here some additions to a fuzzy variable radius niche technique called Dynamic Niche Clustering (DNC) [3, 4 & 5] that enable the identification and creation of niches of arbitrary shape through a mechanism called Niche Linkage. We show that by using this mechanism it is possible to attain better feature extraction from the underlying population.
متن کاملAn Efficient Hierarchical Clustering Technique for Speciation
The sharing method or the basic clearing procedure are reliable niching methods. However, they suffer from a quite high complexity: from O(n2) to O(c n), where c is the number of subpopulations and n is the population size. In this paper, an efficient hierarchical clustering technique is presented, reducing the complexity of these niching methods to O(n log n). On the other hand, the minimal ni...
متن کاملFuzzy particle swarm optimization with nearest-better neighborhood for multimodal optimization
In the last decades, many efforts have been made to solve multimodal optimization problems using Particle Swarm Optimization (PSO). To produce good results, these PSO algorithms need to specify some niching parameters to define the local neighborhood. In this paper, our motivation is to propose the novel neighborhood structures that remove undesirable niching parameters without sacrificing perf...
متن کامل