The incorporation of an efficient initialization method and parameter adaptation using self-organizing maps to solve the TSP

Zhang, Wen Dong; Bai, Yan; Hong, Hui

doi:10.1016/j.amc.2005.02.025

Cited by 10 publications

(6 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…While the system is simple in nature, it requires preprocessing of the input. The authors in [30] propose three approaches to initialization. Although these can be regarded as application specific, they require either preliminary clustering or some form of input space preprocessing.…”

Section: Self-organizationmentioning

confidence: 99%

Fractal initialization for high-quality mapping with self-organizing maps

Valova

Beaton

Buer

et al. 2010

Neural Comput & Applic

View full text Add to dashboard Cite

Initialization of self-organizing maps is typically based on random vectors within the given input space. The implicit problem with random initialization is the overlap (entanglement) of connections between neurons. In this paper, we present a new method of initialization based on a set of self-similar curves known as Hilbert curves. Hilbert curves can be scaled in network size for the number of neurons based on a simple recursive (fractal) technique, implicit in the properties of Hilbert curves. We have shown that when using Hilbert curve vector (HCV) initialization in both classical SOM algorithm and in a parallel-growing algorithm (ParaSOM), the neural network reaches better coverage and faster organization.

show abstract

Section: Self-organizationmentioning

confidence: 99%

Fractal initialization for high-quality mapping with self-organizing maps

Valova

Beaton

Buer

et al. 2010

Neural Comput & Applic

View full text Add to dashboard Cite

show abstract

“…They used initial weights representing nodes on a rectangular frame around the cities, and the authors reported superior results in the selected TSPLIB [31] instances in comparison with the SOM approaches [19,1,9]. These simplified rules have also been applied in [39], where the authors proposed to use l ¼ 1= ffiffiffi k 4 p and the initial value of the gain G 0 = 10. For small values of G, the value of the neighborhood function is very small; thus, the neighboring nodes are negligibly moved.…”

Section: Related Workmentioning

confidence: 94%

“…The rules proposed in [39] and denoted as Zhang-Bai-Hu rules are also used in SME. Furthermore, the original SME adaptation rule is complemented by the decreasing size of the winner node neighborhood, i.e., the size d is updated to d = 0.98d at the end of each adaptation step.…”

Section: Adaptation Parametersmentioning

confidence: 99%

“…In these approaches, the adaptation rules have been modified [37,39], heuristics have been considered [30], and combinations with genetic algorithms [26], memetic [12] or immune system [27] approaches have been proposed. Even though these new approaches improve the performance of SOM for the TSP, SOM is still not competitive with the combinatorial approaches to the TSP in both aspects: the solution quality and required computational time [12].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

On the performance of self-organizing maps for the non-Euclidean Traveling Salesman Problem in the polygonal domain

Faigl

2011

Information Sciences

View full text Add to dashboard Cite

“…O Problema do Caixeiro Viajante (PCV) apresenta-se como um clássico exemplo de Problema de Otimização Combinatória, caracteriza-se por um conjunto de n cidades e a matriz de distâncias entre elas (ZHANG;BAI;HU, 2006). O objetivo é encontrar um ciclo, otimizando a distância a ser percorrida, para que todas as n cidades sejam visitadas uma única vez, retornando à de origem (GOLDBARG;LUNA, 2000).…”

unclassified

Técnicas de pesquisa operacional aplicadas na otimização de rotas de uma rede de lojas de materiais de construção

Moro¹,

Weise²,

Reis³

et al. 2018

Produção em Foco

View full text Add to dashboard Cite

A competitividade tem exigido que as empresas desenvolvam vantagens competitivas em relação aos seus concorrentes que envolvem tempo, custo e nível de serviços. O gerenciamento logístico através de técnicas da pesquisa operacional surge como uma forte ferramenta. Tendo em vista que a distribuição é uma das atividades da logística com maior custo para as empresas, o objetivo do estudo é minimizar a rota de distribuição de materiais de construção, aplicando heurísticas de solução do Problema do Caixeiro Viajante. Para isso utiliza-se como procedimentos técnicos a pesquisa bibliográfica e o estudo de caso. São apresentados definições, formulação e métodos de resolução do PCV através das heurísticas do Vizinho mais Próximo e Subcircuito Inverso, sendo utilizando o software SCILAB® 5.2.2. para a programação dos algoritmos. Em seguida, comparam-se os resultados obtidos através da metodologia com a rota empregada atualmente. Nos resultados alcança-se redução de 8,53% e 15,96% às heurísticas do vizinho mais próximo e Subcircuito Inverso, respectivamente. Por fim, é analisado o custo entre as rotas encontradas, como fator relevante à tomada de decisão.

show abstract

The incorporation of an efficient initialization method and parameter adaptation using self-organizing maps to solve the TSP

Cited by 10 publications

References 12 publications

Fractal initialization for high-quality mapping with self-organizing maps

Fractal initialization for high-quality mapping with self-organizing maps

On the performance of self-organizing maps for the non-Euclidean Traveling Salesman Problem in the polygonal domain

Técnicas de pesquisa operacional aplicadas na otimização de rotas de uma rede de lojas de materiais de construção

Contact Info

Product

Resources

About