Efficient planning through automatic configuration and machine learning

Research output: Contribution to journalArticle

Abstract

Although in the last decade the performance of domain-independent planners has significantly improved, there is no planner that outperforms all the others in every benchmark domains. In many domains, the planning performance can be improved by automatically deriving and exploiting knowledge about the domain and problem structure that is not explicitly given in the input formalization, and that can be used for optimizing the planner behavior. This thesis proposes three innovative techniques for deriving and using additional domain and problem knowledge through automatic algorithm configuration and machine learning techniques: a prediction model of the planning horizons, a method for synthesizing domain-optimized planners from a highly parameterized generic planner, and a system for the automatic configuration of a portfolio planner.

LanguageEnglish
Pages319-321
Number of pages3
JournalAI Communications
Volume26
Issue number3
DOIs
Publication statusPublished - 30 Sep 2013
Externally publishedYes

Fingerprint

Learning systems
Planning

Cite this

@article{3fd4856c60104407b5a0933cc1448767,
title = "Efficient planning through automatic configuration and machine learning",
abstract = "Although in the last decade the performance of domain-independent planners has significantly improved, there is no planner that outperforms all the others in every benchmark domains. In many domains, the planning performance can be improved by automatically deriving and exploiting knowledge about the domain and problem structure that is not explicitly given in the input formalization, and that can be used for optimizing the planner behavior. This thesis proposes three innovative techniques for deriving and using additional domain and problem knowledge through automatic algorithm configuration and machine learning techniques: a prediction model of the planning horizons, a method for synthesizing domain-optimized planners from a highly parameterized generic planner, and a system for the automatic configuration of a portfolio planner.",
keywords = "Automated planning, learning for planning",
author = "Mauro Vallati",
year = "2013",
month = "9",
day = "30",
doi = "10.3233/AIC-130561",
language = "English",
volume = "26",
pages = "319--321",
journal = "AI Communications",
issn = "0921-7126",
publisher = "IOS Press",
number = "3",

}

Efficient planning through automatic configuration and machine learning. / Vallati, Mauro.

In: AI Communications, Vol. 26, No. 3, 30.09.2013, p. 319-321.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Efficient planning through automatic configuration and machine learning

AU - Vallati, Mauro

PY - 2013/9/30

Y1 - 2013/9/30

N2 - Although in the last decade the performance of domain-independent planners has significantly improved, there is no planner that outperforms all the others in every benchmark domains. In many domains, the planning performance can be improved by automatically deriving and exploiting knowledge about the domain and problem structure that is not explicitly given in the input formalization, and that can be used for optimizing the planner behavior. This thesis proposes three innovative techniques for deriving and using additional domain and problem knowledge through automatic algorithm configuration and machine learning techniques: a prediction model of the planning horizons, a method for synthesizing domain-optimized planners from a highly parameterized generic planner, and a system for the automatic configuration of a portfolio planner.

AB - Although in the last decade the performance of domain-independent planners has significantly improved, there is no planner that outperforms all the others in every benchmark domains. In many domains, the planning performance can be improved by automatically deriving and exploiting knowledge about the domain and problem structure that is not explicitly given in the input formalization, and that can be used for optimizing the planner behavior. This thesis proposes three innovative techniques for deriving and using additional domain and problem knowledge through automatic algorithm configuration and machine learning techniques: a prediction model of the planning horizons, a method for synthesizing domain-optimized planners from a highly parameterized generic planner, and a system for the automatic configuration of a portfolio planner.

KW - Automated planning

KW - learning for planning

UR - http://www.scopus.com/inward/record.url?scp=84884536850&partnerID=8YFLogxK

U2 - 10.3233/AIC-130561

DO - 10.3233/AIC-130561

M3 - Article

VL - 26

SP - 319

EP - 321

JO - AI Communications

T2 - AI Communications

JF - AI Communications

SN - 0921-7126

IS - 3

ER -