Loading…

Using Meta-learning to Recommend Meta-heuristics for the Traveling Salesman Problem

Several optimization methods can find good solutions for different instances of the Traveling Salesman Problem (TSP). Since there is no method that generates the best solution for all instances, the selection of the most promising method for a given TSP instance is a difficult task. This paper descr...

Full description

Saved in:
Bibliographic Details
Main Authors: Kanda, J. Y., de Carvalho, A. C. P. L. F., Hruschka, E. R., Soares, C.
Format: Conference Proceeding
Language:English
Subjects:
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
cited_by
cites
container_end_page 351
container_issue
container_start_page 346
container_title
container_volume 1
creator Kanda, J. Y.
de Carvalho, A. C. P. L. F.
Hruschka, E. R.
Soares, C.
description Several optimization methods can find good solutions for different instances of the Traveling Salesman Problem (TSP). Since there is no method that generates the best solution for all instances, the selection of the most promising method for a given TSP instance is a difficult task. This paper describes a meta-learning-based approach to select optimization methods for the TSP. Multilayer perceptron (MLP) networks are trained with TSP examples. These examples are described by a set of TSP characteristics and the cost of solutions obtained by a set of optimization methods. The trained MLP network model is then used to predict a ranking of these methods for a new TSP instance. Correlation measures are used to compare the predicted ranking with the ranking previously known. The obtained results suggest that the proposed approach is promising.
doi_str_mv 10.1109/ICMLA.2011.153
format conference_proceeding
fullrecord <record><control><sourceid>ieee_6IE</sourceid><recordid>TN_cdi_ieee_primary_6146996</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><ieee_id>6146996</ieee_id><sourcerecordid>6146996</sourcerecordid><originalsourceid>FETCH-LOGICAL-i90t-ee0bd22b05941cc16c83150c4d026bcce9e4b622a939278c87079151448ede323</originalsourceid><addsrcrecordid>eNotj01Lw0AYhFdEUGuuXrzsH0jcdz-zxxL8KKQoNp7LZvPWruRDdqPgv7elzmUYnmFgCLkFVgAwe7-q1vWy4AygACXOyDUz2iqpmeHnJLOmBKmM4SAkvyRZSp_sIK2tBXNFNu8pjB90jbPLe3RxPKZ5om_op2HAsTuhPX7HkObgE91Nkc57pE10P9gf6xvXYxrcSF_j1PY43JCLnesTZv--IM3jQ1M95_XL06pa1nmwbM4RWdtx3jJlJXgP2pcCFPOyY1y33qNF2WrOnRWWm9KXhhkLCqQssUPBxYLcnWYDIm6_Yhhc_N1qkIdrWvwBVTJP3Q</addsrcrecordid><sourcetype>Publisher</sourcetype><iscdi>true</iscdi><recordtype>conference_proceeding</recordtype></control><display><type>conference_proceeding</type><title>Using Meta-learning to Recommend Meta-heuristics for the Traveling Salesman Problem</title><source>IEEE Electronic Library (IEL) Conference Proceedings</source><creator>Kanda, J. Y. ; de Carvalho, A. C. P. L. F. ; Hruschka, E. R. ; Soares, C.</creator><creatorcontrib>Kanda, J. Y. ; de Carvalho, A. C. P. L. F. ; Hruschka, E. R. ; Soares, C.</creatorcontrib><description>Several optimization methods can find good solutions for different instances of the Traveling Salesman Problem (TSP). Since there is no method that generates the best solution for all instances, the selection of the most promising method for a given TSP instance is a difficult task. This paper describes a meta-learning-based approach to select optimization methods for the TSP. Multilayer perceptron (MLP) networks are trained with TSP examples. These examples are described by a set of TSP characteristics and the cost of solutions obtained by a set of optimization methods. The trained MLP network model is then used to predict a ranking of these methods for a new TSP instance. Correlation measures are used to compare the predicted ranking with the ranking previously known. The obtained results suggest that the proposed approach is promising.</description><identifier>ISBN: 9781457721342</identifier><identifier>ISBN: 1457721341</identifier><identifier>EISBN: 0769546072</identifier><identifier>EISBN: 9780769546070</identifier><identifier>DOI: 10.1109/ICMLA.2011.153</identifier><language>eng</language><publisher>IEEE</publisher><subject>Algorithm selection ; Cities and towns ; Genetic algorithms ; meta-learning ; multilayer perceptron network ; Neurons ; Optimization methods ; Prediction algorithms ; Predictive models ; traveling salesman problem</subject><ispartof>2011 10th International Conference on Machine Learning and Applications and Workshops, 2011, Vol.1, p.346-351</ispartof><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktohtml>$$Uhttps://ieeexplore.ieee.org/document/6146996$$EHTML$$P50$$Gieee$$H</linktohtml><link.rule.ids>309,310,778,782,787,788,2054,27908,54903</link.rule.ids><linktorsrc>$$Uhttps://ieeexplore.ieee.org/document/6146996$$EView_record_in_IEEE$$FView_record_in_$$GIEEE</linktorsrc></links><search><creatorcontrib>Kanda, J. Y.</creatorcontrib><creatorcontrib>de Carvalho, A. C. P. L. F.</creatorcontrib><creatorcontrib>Hruschka, E. R.</creatorcontrib><creatorcontrib>Soares, C.</creatorcontrib><title>Using Meta-learning to Recommend Meta-heuristics for the Traveling Salesman Problem</title><title>2011 10th International Conference on Machine Learning and Applications and Workshops</title><addtitle>icmla</addtitle><description>Several optimization methods can find good solutions for different instances of the Traveling Salesman Problem (TSP). Since there is no method that generates the best solution for all instances, the selection of the most promising method for a given TSP instance is a difficult task. This paper describes a meta-learning-based approach to select optimization methods for the TSP. Multilayer perceptron (MLP) networks are trained with TSP examples. These examples are described by a set of TSP characteristics and the cost of solutions obtained by a set of optimization methods. The trained MLP network model is then used to predict a ranking of these methods for a new TSP instance. Correlation measures are used to compare the predicted ranking with the ranking previously known. The obtained results suggest that the proposed approach is promising.</description><subject>Algorithm selection</subject><subject>Cities and towns</subject><subject>Genetic algorithms</subject><subject>meta-learning</subject><subject>multilayer perceptron network</subject><subject>Neurons</subject><subject>Optimization methods</subject><subject>Prediction algorithms</subject><subject>Predictive models</subject><subject>traveling salesman problem</subject><isbn>9781457721342</isbn><isbn>1457721341</isbn><isbn>0769546072</isbn><isbn>9780769546070</isbn><fulltext>true</fulltext><rsrctype>conference_proceeding</rsrctype><creationdate>2011</creationdate><recordtype>conference_proceeding</recordtype><sourceid>6IE</sourceid><recordid>eNotj01Lw0AYhFdEUGuuXrzsH0jcdz-zxxL8KKQoNp7LZvPWruRDdqPgv7elzmUYnmFgCLkFVgAwe7-q1vWy4AygACXOyDUz2iqpmeHnJLOmBKmM4SAkvyRZSp_sIK2tBXNFNu8pjB90jbPLe3RxPKZ5om_op2HAsTuhPX7HkObgE91Nkc57pE10P9gf6xvXYxrcSF_j1PY43JCLnesTZv--IM3jQ1M95_XL06pa1nmwbM4RWdtx3jJlJXgP2pcCFPOyY1y33qNF2WrOnRWWm9KXhhkLCqQssUPBxYLcnWYDIm6_Yhhc_N1qkIdrWvwBVTJP3Q</recordid><startdate>201112</startdate><enddate>201112</enddate><creator>Kanda, J. Y.</creator><creator>de Carvalho, A. C. P. L. F.</creator><creator>Hruschka, E. R.</creator><creator>Soares, C.</creator><general>IEEE</general><scope>6IE</scope><scope>6IL</scope><scope>CBEJK</scope><scope>RIE</scope><scope>RIL</scope></search><sort><creationdate>201112</creationdate><title>Using Meta-learning to Recommend Meta-heuristics for the Traveling Salesman Problem</title><author>Kanda, J. Y. ; de Carvalho, A. C. P. L. F. ; Hruschka, E. R. ; Soares, C.</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-i90t-ee0bd22b05941cc16c83150c4d026bcce9e4b622a939278c87079151448ede323</frbrgroupid><rsrctype>conference_proceedings</rsrctype><prefilter>conference_proceedings</prefilter><language>eng</language><creationdate>2011</creationdate><topic>Algorithm selection</topic><topic>Cities and towns</topic><topic>Genetic algorithms</topic><topic>meta-learning</topic><topic>multilayer perceptron network</topic><topic>Neurons</topic><topic>Optimization methods</topic><topic>Prediction algorithms</topic><topic>Predictive models</topic><topic>traveling salesman problem</topic><toplevel>online_resources</toplevel><creatorcontrib>Kanda, J. Y.</creatorcontrib><creatorcontrib>de Carvalho, A. C. P. L. F.</creatorcontrib><creatorcontrib>Hruschka, E. R.</creatorcontrib><creatorcontrib>Soares, C.</creatorcontrib><collection>IEEE Electronic Library (IEL) Conference Proceedings</collection><collection>IEEE Proceedings Order Plan All Online (POP All Online) 1998-present by volume</collection><collection>IEEE Xplore All Conference Proceedings</collection><collection>IEEE Electronic Library (IEL)</collection><collection>IEEE Proceedings Order Plans (POP All) 1998-Present</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Kanda, J. Y.</au><au>de Carvalho, A. C. P. L. F.</au><au>Hruschka, E. R.</au><au>Soares, C.</au><format>book</format><genre>proceeding</genre><ristype>CONF</ristype><atitle>Using Meta-learning to Recommend Meta-heuristics for the Traveling Salesman Problem</atitle><btitle>2011 10th International Conference on Machine Learning and Applications and Workshops</btitle><stitle>icmla</stitle><date>2011-12</date><risdate>2011</risdate><volume>1</volume><spage>346</spage><epage>351</epage><pages>346-351</pages><isbn>9781457721342</isbn><isbn>1457721341</isbn><eisbn>0769546072</eisbn><eisbn>9780769546070</eisbn><abstract>Several optimization methods can find good solutions for different instances of the Traveling Salesman Problem (TSP). Since there is no method that generates the best solution for all instances, the selection of the most promising method for a given TSP instance is a difficult task. This paper describes a meta-learning-based approach to select optimization methods for the TSP. Multilayer perceptron (MLP) networks are trained with TSP examples. These examples are described by a set of TSP characteristics and the cost of solutions obtained by a set of optimization methods. The trained MLP network model is then used to predict a ranking of these methods for a new TSP instance. Correlation measures are used to compare the predicted ranking with the ranking previously known. The obtained results suggest that the proposed approach is promising.</abstract><pub>IEEE</pub><doi>10.1109/ICMLA.2011.153</doi><tpages>6</tpages></addata></record>
fulltext fulltext_linktorsrc
identifier ISBN: 9781457721342
ispartof 2011 10th International Conference on Machine Learning and Applications and Workshops, 2011, Vol.1, p.346-351
issn
language eng
recordid cdi_ieee_primary_6146996
source IEEE Electronic Library (IEL) Conference Proceedings
subjects Algorithm selection
Cities and towns
Genetic algorithms
meta-learning
multilayer perceptron network
Neurons
Optimization methods
Prediction algorithms
Predictive models
traveling salesman problem
title Using Meta-learning to Recommend Meta-heuristics for the Traveling Salesman Problem
url http://sfxeu10.hosted.exlibrisgroup.com/loughborough?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-16T19%3A12%3A38IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-ieee_6IE&rft_val_fmt=info:ofi/fmt:kev:mtx:book&rft.genre=proceeding&rft.atitle=Using%20Meta-learning%20to%20Recommend%20Meta-heuristics%20for%20the%20Traveling%20Salesman%20Problem&rft.btitle=2011%2010th%20International%20Conference%20on%20Machine%20Learning%20and%20Applications%20and%20Workshops&rft.au=Kanda,%20J.%20Y.&rft.date=2011-12&rft.volume=1&rft.spage=346&rft.epage=351&rft.pages=346-351&rft.isbn=9781457721342&rft.isbn_list=1457721341&rft_id=info:doi/10.1109/ICMLA.2011.153&rft.eisbn=0769546072&rft.eisbn_list=9780769546070&rft_dat=%3Cieee_6IE%3E6146996%3C/ieee_6IE%3E%3Cgrp_id%3Ecdi_FETCH-LOGICAL-i90t-ee0bd22b05941cc16c83150c4d026bcce9e4b622a939278c87079151448ede323%3C/grp_id%3E%3Coa%3E%3C/oa%3E%3Curl%3E%3C/url%3E&rft_id=info:oai/&rft_id=info:pmid/&rft_ieee_id=6146996&rfr_iscdi=true