On the predictability of domain-independent temporal planners

Isabel Cenamor, Mauro Vallati, Lukáš Chrpa

Research output: Contribution to journalArticle

Abstract

Temporal planning is a research discipline that addresses the problem of generating a totally or a partially ordered sequence of actions that transform the environment from some initial state to a desired goal state, while taking into account time constraints and actions' duration. For its ability to describe and address temporal constraints, temporal planning is of critical importance for a wide range of real-world applications. Predicting the performance of temporal planners can lead to significant improvements in the area, as planners can then be combined in order to boost the performance on a given set of problem instances. This paper investigates the predictability of the state-of-the-art temporal planners by introducing a new set of temporal-specific features and exploiting them for generating classification and regression empirical performance models (EPMs) of considered planners. EPMs are also tested with regard to their ability to select the most promising planner for efficiently solving a given temporal planning problem. Our extensive empirical analysis indicates that the introduced set of features allows to generate EPMs that can effectively perform algorithm selection, and the use of EPMs is therefore a promising direction for improving the state of the art of temporal planning, hence fostering the use of planning in real-world applications.

LanguageEnglish
Number of pages29
JournalComputational Intelligence
Early online date22 Apr 2019
DOIs
Publication statusE-pub ahead of print - 22 Apr 2019

Fingerprint

Predictability
Empirical Model
Performance Model
Planning
Real-world Applications
Temporal Constraints
Empirical Analysis
Regression
Transform
Range of data

Cite this

@article{533841f5b57d4b24b26260771ce81ee8,
title = "On the predictability of domain-independent temporal planners",
abstract = "Temporal planning is a research discipline that addresses the problem of generating a totally or a partially ordered sequence of actions that transform the environment from some initial state to a desired goal state, while taking into account time constraints and actions' duration. For its ability to describe and address temporal constraints, temporal planning is of critical importance for a wide range of real-world applications. Predicting the performance of temporal planners can lead to significant improvements in the area, as planners can then be combined in order to boost the performance on a given set of problem instances. This paper investigates the predictability of the state-of-the-art temporal planners by introducing a new set of temporal-specific features and exploiting them for generating classification and regression empirical performance models (EPMs) of considered planners. EPMs are also tested with regard to their ability to select the most promising planner for efficiently solving a given temporal planning problem. Our extensive empirical analysis indicates that the introduced set of features allows to generate EPMs that can effectively perform algorithm selection, and the use of EPMs is therefore a promising direction for improving the state of the art of temporal planning, hence fostering the use of planning in real-world applications.",
keywords = "Automated planning, Predicting performance, Temporal planning",
author = "Isabel Cenamor and Mauro Vallati and Luk{\'a}š Chrpa",
year = "2019",
month = "4",
day = "22",
doi = "10.1111/coin.12211",
language = "English",
journal = "Computational Intelligence",
issn = "0824-7935",
publisher = "Wiley-Blackwell",

}

On the predictability of domain-independent temporal planners. / Cenamor, Isabel; Vallati, Mauro; Chrpa, Lukáš.

In: Computational Intelligence, 22.04.2019.

Research output: Contribution to journalArticle

TY - JOUR

T1 - On the predictability of domain-independent temporal planners

AU - Cenamor, Isabel

AU - Vallati, Mauro

AU - Chrpa, Lukáš

PY - 2019/4/22

Y1 - 2019/4/22

N2 - Temporal planning is a research discipline that addresses the problem of generating a totally or a partially ordered sequence of actions that transform the environment from some initial state to a desired goal state, while taking into account time constraints and actions' duration. For its ability to describe and address temporal constraints, temporal planning is of critical importance for a wide range of real-world applications. Predicting the performance of temporal planners can lead to significant improvements in the area, as planners can then be combined in order to boost the performance on a given set of problem instances. This paper investigates the predictability of the state-of-the-art temporal planners by introducing a new set of temporal-specific features and exploiting them for generating classification and regression empirical performance models (EPMs) of considered planners. EPMs are also tested with regard to their ability to select the most promising planner for efficiently solving a given temporal planning problem. Our extensive empirical analysis indicates that the introduced set of features allows to generate EPMs that can effectively perform algorithm selection, and the use of EPMs is therefore a promising direction for improving the state of the art of temporal planning, hence fostering the use of planning in real-world applications.

AB - Temporal planning is a research discipline that addresses the problem of generating a totally or a partially ordered sequence of actions that transform the environment from some initial state to a desired goal state, while taking into account time constraints and actions' duration. For its ability to describe and address temporal constraints, temporal planning is of critical importance for a wide range of real-world applications. Predicting the performance of temporal planners can lead to significant improvements in the area, as planners can then be combined in order to boost the performance on a given set of problem instances. This paper investigates the predictability of the state-of-the-art temporal planners by introducing a new set of temporal-specific features and exploiting them for generating classification and regression empirical performance models (EPMs) of considered planners. EPMs are also tested with regard to their ability to select the most promising planner for efficiently solving a given temporal planning problem. Our extensive empirical analysis indicates that the introduced set of features allows to generate EPMs that can effectively perform algorithm selection, and the use of EPMs is therefore a promising direction for improving the state of the art of temporal planning, hence fostering the use of planning in real-world applications.

KW - Automated planning

KW - Predicting performance

KW - Temporal planning

UR - http://www.scopus.com/inward/record.url?scp=85064605442&partnerID=8YFLogxK

U2 - 10.1111/coin.12211

DO - 10.1111/coin.12211

M3 - Article

JO - Computational Intelligence

T2 - Computational Intelligence

JF - Computational Intelligence

SN - 0824-7935

ER -