A Meta-Learning Approach to Select Meta-Heuristics for the Traveling Salesman Problem Using MLP-Based Label Ranking

被引：0

作者：

Kanda, Jorge ^{[1
,2
]}

Soares, Carlos ^{[3
,4
]}

Hruschka, Eduardo ^{[1
]}

de Carvalho, Andre ^{[1
]}

机构：

[1] Univ Sao Paulo, Inst Ciencias Matemat & Computacao, Ave Trabalhador Sao Carlense 400, BR-13566590 Sao Carlos, SP, Brazil

[2] Univ Fed Amazonas, Inst Ciencias Exatas Tecnol, BR-69103128 Itacoatiara, Brazil

[3] Univ Porto, Fac Econ, P-4200464 Oporto, Portugal

[4] INESC, TEC Porto LA, P-4200464 Oporto, Portugal

来源：

NEURAL INFORMATION PROCESSING, ICONIP 2012, PT III | 2012年 / 7665卷

关键词：

meta-learning; label ranking; multilayer perceptron; traveling salesman problem; OPTIMIZATION;

D O I：

暂无

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

Different meta-heuristics (MHs) may find the best solutions for different traveling salesman problem (TSP) instances. The a priori selection of the best MH for a given instance is a difficult task. We address this task by using a meta-learning based approach, which ranks different MHs according to their expected performance. Our approach uses Multilayer Perceptrons (MLPs) for label ranking. It is tested on two different TSP scenarios, namely: re-visiting customers and visiting prospects. The experimental results show that: 1) MLPs can accurately predict MH rankings for TSP, 2) better TSP solutions can be obtained from a label ranking compared to multilabel classification approach, and 3) it is important to consider different TSP application scenarios when using meta-learning for MH selection.

引用

页码：488 / 495

页数：8

共 25 条

[21] The proof and measurement of association between two things
Spearman, C
[J]. AMERICAN JOURNAL OF PSYCHOLOGY, 1904, 15 : 72 - 101
[22] Vembu S, 2010, PREFERENCE LEARNING, P45, DOI 10.1007/978-3-642-14125-6_3
[23] A perspective view and survey of meta-learning
Vilalta, R
Drissi, Y
[J]. ARTIFICIAL INTELLIGENCE REVIEW, 2002, 18 (02) : 77 - 95
[24] Wolpert D. H., 1997, IEEE Transactions on Evolutionary Computation, V1, P67, DOI 10.1109/4235.585893
[25] [No title captured]

← 1 2 3 →