Portfolio Methods for Optimal Planning: An Empirical Analysis

Mattia Rizzini, Chris Fawcett, Mauro Vallati, Alfonso E. Gerevini, Holger H. Hoos

Research output: Chapter in Book/Report/Conference proceedingConference contribution

3 Citations (Scopus)

Abstract

Combining the complementary strengths of several algorithms through portfolio approaches has been demonstrated to be effective in solving a wide range of AI problems. Notably, portfolio techniques have been prominently applied to suboptimal (satisficing) AI planning. Here, we consider the construction of sequential planner portfolios for (domain-independent) optimal planning. Specifically, we introduce four techniques (three of which are dynamic) for per-instance planner schedule generation using problem instance features, and investigate the usefulness of a range of static and dynamic techniques for combining planners. Our extensive experimental analysis demonstrates the benefits of using static and dynamic sequential portfolios for optimal planning, and provides insights on the most suitable conditions for their fruitful exploitation.
LanguageEnglish
Title of host publication2015 IEEE 27th International Conference on Tools with Artificial Intelligence
PublisherIEEE
Pages494-501
Number of pages8
ISBN (Electronic)9781509001637, 9781509001620
DOIs
Publication statusPublished - 7 Jan 2016
Event2015 IEEE 27th International Conference on Tools with Artificial Intelligence - Vietri sul Mare, Italy
Duration: 9 Nov 201511 Nov 2015
Conference number: 27

Conference

Conference2015 IEEE 27th International Conference on Tools with Artificial Intelligence
Abbreviated titleICTAI
CountryItaly
CityVietri sul Mare
Period9/11/1511/11/15

Fingerprint

Planning

Cite this

Rizzini, M., Fawcett, C., Vallati, M., Gerevini, A. E., & Hoos, H. H. (2016). Portfolio Methods for Optimal Planning: An Empirical Analysis. In 2015 IEEE 27th International Conference on Tools with Artificial Intelligence (pp. 494-501). IEEE. https://doi.org/10.1109/ICTAI.2015.79
Rizzini, Mattia ; Fawcett, Chris ; Vallati, Mauro ; Gerevini, Alfonso E. ; Hoos, Holger H. / Portfolio Methods for Optimal Planning : An Empirical Analysis. 2015 IEEE 27th International Conference on Tools with Artificial Intelligence . IEEE, 2016. pp. 494-501
@inproceedings{a4b9ec3515db410f8d5c89241a78bcf7,
title = "Portfolio Methods for Optimal Planning: An Empirical Analysis",
abstract = "Combining the complementary strengths of several algorithms through portfolio approaches has been demonstrated to be effective in solving a wide range of AI problems. Notably, portfolio techniques have been prominently applied to suboptimal (satisficing) AI planning. Here, we consider the construction of sequential planner portfolios for (domain-independent) optimal planning. Specifically, we introduce four techniques (three of which are dynamic) for per-instance planner schedule generation using problem instance features, and investigate the usefulness of a range of static and dynamic techniques for combining planners. Our extensive experimental analysis demonstrates the benefits of using static and dynamic sequential portfolios for optimal planning, and provides insights on the most suitable conditions for their fruitful exploitation.",
keywords = "portfolios, runtime, training, Feature extraction, schedules, Artificial intelligence, Automated Planning, optimal planning, sequential portfolio, per-instance portfolio generation",
author = "Mattia Rizzini and Chris Fawcett and Mauro Vallati and Gerevini, {Alfonso E.} and Hoos, {Holger H.}",
year = "2016",
month = "1",
day = "7",
doi = "10.1109/ICTAI.2015.79",
language = "English",
pages = "494--501",
booktitle = "2015 IEEE 27th International Conference on Tools with Artificial Intelligence",
publisher = "IEEE",

}

Rizzini, M, Fawcett, C, Vallati, M, Gerevini, AE & Hoos, HH 2016, Portfolio Methods for Optimal Planning: An Empirical Analysis. in 2015 IEEE 27th International Conference on Tools with Artificial Intelligence . IEEE, pp. 494-501, 2015 IEEE 27th International Conference on Tools with Artificial Intelligence, Vietri sul Mare, Italy, 9/11/15. https://doi.org/10.1109/ICTAI.2015.79

Portfolio Methods for Optimal Planning : An Empirical Analysis. / Rizzini, Mattia; Fawcett, Chris; Vallati, Mauro; Gerevini, Alfonso E.; Hoos, Holger H.

2015 IEEE 27th International Conference on Tools with Artificial Intelligence . IEEE, 2016. p. 494-501.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - Portfolio Methods for Optimal Planning

T2 - An Empirical Analysis

AU - Rizzini, Mattia

AU - Fawcett, Chris

AU - Vallati, Mauro

AU - Gerevini, Alfonso E.

AU - Hoos, Holger H.

PY - 2016/1/7

Y1 - 2016/1/7

N2 - Combining the complementary strengths of several algorithms through portfolio approaches has been demonstrated to be effective in solving a wide range of AI problems. Notably, portfolio techniques have been prominently applied to suboptimal (satisficing) AI planning. Here, we consider the construction of sequential planner portfolios for (domain-independent) optimal planning. Specifically, we introduce four techniques (three of which are dynamic) for per-instance planner schedule generation using problem instance features, and investigate the usefulness of a range of static and dynamic techniques for combining planners. Our extensive experimental analysis demonstrates the benefits of using static and dynamic sequential portfolios for optimal planning, and provides insights on the most suitable conditions for their fruitful exploitation.

AB - Combining the complementary strengths of several algorithms through portfolio approaches has been demonstrated to be effective in solving a wide range of AI problems. Notably, portfolio techniques have been prominently applied to suboptimal (satisficing) AI planning. Here, we consider the construction of sequential planner portfolios for (domain-independent) optimal planning. Specifically, we introduce four techniques (three of which are dynamic) for per-instance planner schedule generation using problem instance features, and investigate the usefulness of a range of static and dynamic techniques for combining planners. Our extensive experimental analysis demonstrates the benefits of using static and dynamic sequential portfolios for optimal planning, and provides insights on the most suitable conditions for their fruitful exploitation.

KW - portfolios

KW - runtime

KW - training

KW - Feature extraction

KW - schedules

KW - Artificial intelligence

KW - Automated Planning

KW - optimal planning

KW - sequential portfolio

KW - per-instance portfolio generation

U2 - 10.1109/ICTAI.2015.79

DO - 10.1109/ICTAI.2015.79

M3 - Conference contribution

SP - 494

EP - 501

BT - 2015 IEEE 27th International Conference on Tools with Artificial Intelligence

PB - IEEE

ER -

Rizzini M, Fawcett C, Vallati M, Gerevini AE, Hoos HH. Portfolio Methods for Optimal Planning: An Empirical Analysis. In 2015 IEEE 27th International Conference on Tools with Artificial Intelligence . IEEE. 2016. p. 494-501 https://doi.org/10.1109/ICTAI.2015.79