ASAP: An automatic algorithm selection approach for planning

Research output: Contribution to journalArticle

4 Citations (Scopus)

Abstract

Despite the advances made in the last decade in automated planning, no planner outperforms all the others in every known benchmark domain. This observation motivates the idea of selecting different planning algorithms for different domains. Moreover, the planners' performances are affected by the structure of the search space, which depends on the encoding of the considered domain. In many domains, the performance of a planner can be improved by exploiting additional knowledge, for instance, in the form of macro-operators or entanglements. In this paper we propose ASAP, an automatic Algorithm Selection Approach for Planning that: (i) for a given domain initially learns additional knowledge, in the form of macro-operators and entanglements, which is used for creating different encodings of the given planning domain and problems, and (ii) explores the 2 dimensional space of available algorithms, defined as encodings-planners couples, and then (iii) selects the most promising algorithm for optimising either the runtimes or the quality of the solution plans.

LanguageEnglish
Article number1460032
JournalInternational Journal on Artificial Intelligence Tools
Volume23
Issue number6
DOIs
Publication statusPublished - 29 Dec 2014

Fingerprint

Planning
Macros

Cite this

@article{4325e7c00798435393ffa1cd221826b8,
title = "ASAP: An automatic algorithm selection approach for planning",
abstract = "Despite the advances made in the last decade in automated planning, no planner outperforms all the others in every known benchmark domain. This observation motivates the idea of selecting different planning algorithms for different domains. Moreover, the planners' performances are affected by the structure of the search space, which depends on the encoding of the considered domain. In many domains, the performance of a planner can be improved by exploiting additional knowledge, for instance, in the form of macro-operators or entanglements. In this paper we propose ASAP, an automatic Algorithm Selection Approach for Planning that: (i) for a given domain initially learns additional knowledge, in the form of macro-operators and entanglements, which is used for creating different encodings of the given planning domain and problems, and (ii) explores the 2 dimensional space of available algorithms, defined as encodings-planners couples, and then (iii) selects the most promising algorithm for optimising either the runtimes or the quality of the solution plans.",
keywords = "algorithm selection, Automated planning, learning for planning",
author = "Mauro Vallati and Luk{\'a}š Chrpa and Diane Kitchin",
year = "2014",
month = "12",
day = "29",
doi = "10.1142/S021821301460032X",
language = "English",
volume = "23",
journal = "International Journal on Artificial Intelligence Tools",
issn = "0218-2130",
publisher = "World Scientific Publishing Co. Pte Ltd",
number = "6",

}

ASAP : An automatic algorithm selection approach for planning. / Vallati, Mauro; Chrpa, Lukáš; Kitchin, Diane.

In: International Journal on Artificial Intelligence Tools, Vol. 23, No. 6, 1460032, 29.12.2014.

Research output: Contribution to journalArticle

TY - JOUR

T1 - ASAP

T2 - International Journal on Artificial Intelligence Tools

AU - Vallati, Mauro

AU - Chrpa, Lukáš

AU - Kitchin, Diane

PY - 2014/12/29

Y1 - 2014/12/29

N2 - Despite the advances made in the last decade in automated planning, no planner outperforms all the others in every known benchmark domain. This observation motivates the idea of selecting different planning algorithms for different domains. Moreover, the planners' performances are affected by the structure of the search space, which depends on the encoding of the considered domain. In many domains, the performance of a planner can be improved by exploiting additional knowledge, for instance, in the form of macro-operators or entanglements. In this paper we propose ASAP, an automatic Algorithm Selection Approach for Planning that: (i) for a given domain initially learns additional knowledge, in the form of macro-operators and entanglements, which is used for creating different encodings of the given planning domain and problems, and (ii) explores the 2 dimensional space of available algorithms, defined as encodings-planners couples, and then (iii) selects the most promising algorithm for optimising either the runtimes or the quality of the solution plans.

AB - Despite the advances made in the last decade in automated planning, no planner outperforms all the others in every known benchmark domain. This observation motivates the idea of selecting different planning algorithms for different domains. Moreover, the planners' performances are affected by the structure of the search space, which depends on the encoding of the considered domain. In many domains, the performance of a planner can be improved by exploiting additional knowledge, for instance, in the form of macro-operators or entanglements. In this paper we propose ASAP, an automatic Algorithm Selection Approach for Planning that: (i) for a given domain initially learns additional knowledge, in the form of macro-operators and entanglements, which is used for creating different encodings of the given planning domain and problems, and (ii) explores the 2 dimensional space of available algorithms, defined as encodings-planners couples, and then (iii) selects the most promising algorithm for optimising either the runtimes or the quality of the solution plans.

KW - algorithm selection

KW - Automated planning

KW - learning for planning

UR - http://www.scopus.com/inward/record.url?scp=84930157324&partnerID=8YFLogxK

U2 - 10.1142/S021821301460032X

DO - 10.1142/S021821301460032X

M3 - Article

VL - 23

JO - International Journal on Artificial Intelligence Tools

JF - International Journal on Artificial Intelligence Tools

SN - 0218-2130

IS - 6

M1 - 1460032

ER -