Mining useful macro-actions in planning

Sandra Castellanos-Paez; Damien Pellier; Humbert Fiorino; Sylvie Pesty

doi:10.1109/ICAIPR.2016.7585227

Mining useful macro-actions in planning

Sandra Castellanos-Paez, Damien Pellier, Humbert Fiorino, Sylvie Pesty

Grenoble Alps University

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

3 Citations (Scopus)

Abstract

Planning has achieved significant progress in recent years. Among the various approaches to scale up plan synthesis, the use of macro-actions has been widely explored. As a first stage towards the development of a solution to learn on-line macro-actions, we propose an algorithm to identify useful macroactions based on data mining techniques. The integration in the planning search of these learned macro-actions shows significant improvements over six classical planning benchmarks.

Original language	English
Title of host publication	2016 Third International Conference on Artificial Intelligence and Pattern Recognition (AIPR)
Publisher	IEEE
Number of pages	6
ISBN (Electronic)	9781467391870, 9781467391863
ISBN (Print)	9781467391887
DOIs	https://doi.org/10.1109/ICAIPR.2016.7585227
Publication status	Published - 13 Oct 2016
Externally published	Yes
Event	3rd International Conference on Artificial Intelligence and Pattern Recognition - Lodz University of Technology, Lodz, Poland Duration: 19 Sept 2016 → 21 Sept 2016 https://cps-vo.org/node/23143

Conference

Conference	3rd International Conference on Artificial Intelligence and Pattern Recognition
Abbreviated title	AIPR2016
Country/Territory	Poland
City	Lodz
Period	19/09/16 → 21/09/16
Internet address	https://cps-vo.org/node/23143

Access to Document

10.1109/ICAIPR.2016.7585227Licence: Unspecified

Cite this

@inproceedings{7ece11644c3147ee9e8182eff5283088,

title = "Mining useful macro-actions in planning",

abstract = "Planning has achieved significant progress in recent years. Among the various approaches to scale up plan synthesis, the use of macro-actions has been widely explored. As a first stage towards the development of a solution to learn on-line macro-actions, we propose an algorithm to identify useful macroactions based on data mining techniques. The integration in the planning search of these learned macro-actions shows significant improvements over six classical planning benchmarks.",

keywords = "automated planning, macro actions, pattern mining, data mining, learning",

author = "Sandra Castellanos-Paez and Damien Pellier and Humbert Fiorino and Sylvie Pesty",

year = "2016",

month = oct,

day = "13",

doi = "10.1109/ICAIPR.2016.7585227",

language = "English",

isbn = "9781467391887",

booktitle = "2016 Third International Conference on Artificial Intelligence and Pattern Recognition (AIPR)",

publisher = "IEEE",

address = "United States",

note = "3rd International Conference on Artificial Intelligence and Pattern Recognition, AIPR2016 ; Conference date: 19-09-2016 Through 21-09-2016",

url = "https://cps-vo.org/node/23143",

}

TY - GEN

T1 - Mining useful macro-actions in planning

AU - Castellanos-Paez, Sandra

AU - Pellier, Damien

AU - Fiorino, Humbert

AU - Pesty, Sylvie

PY - 2016/10/13

Y1 - 2016/10/13

N2 - Planning has achieved significant progress in recent years. Among the various approaches to scale up plan synthesis, the use of macro-actions has been widely explored. As a first stage towards the development of a solution to learn on-line macro-actions, we propose an algorithm to identify useful macroactions based on data mining techniques. The integration in the planning search of these learned macro-actions shows significant improvements over six classical planning benchmarks.

AB - Planning has achieved significant progress in recent years. Among the various approaches to scale up plan synthesis, the use of macro-actions has been widely explored. As a first stage towards the development of a solution to learn on-line macro-actions, we propose an algorithm to identify useful macroactions based on data mining techniques. The integration in the planning search of these learned macro-actions shows significant improvements over six classical planning benchmarks.

KW - automated planning

KW - macro actions

KW - pattern mining

KW - data mining

KW - learning

U2 - 10.1109/ICAIPR.2016.7585227

DO - 10.1109/ICAIPR.2016.7585227

M3 - Conference contribution

SN - 9781467391887

BT - 2016 Third International Conference on Artificial Intelligence and Pattern Recognition (AIPR)

PB - IEEE

T2 - 3rd International Conference on Artificial Intelligence and Pattern Recognition

Y2 - 19 September 2016 through 21 September 2016

ER -

Mining useful macro-actions in planning

Abstract

Conference

Access to Document

Fingerprint

Cite this