ERA: Extracting Planning Macro-Operators from Adjacent and Non-adjacent Sequences

Sandra Castellanos-Paez; Romain Rombourg; Philippe Lalanda

doi:10.1007/978-3-030-69886-7_3

ERA: Extracting Planning Macro-Operators from Adjacent and Non-adjacent Sequences

Sandra Castellanos-Paez, Romain Rombourg, Philippe Lalanda

Grenoble Alps University

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

1 Citation (Scopus)

Abstract

Intuitively, Automated Planning systems capable of learning from previous experiences should be able to achieve better performance. One way to build on past experiences is to augment domains with macro-operators (i.e. frequent operator sequences). In most existing works, macros are generated from chunks of adjacent operators extracted from a set of plans. Although they provide some interesting results this type of analysis may provide incomplete results. In this paper, we propose ERA, an automatic extraction method for macro-operators from a set of solution plans. Our algorithm is domain and planner independent and can find all macro-operator occurrences even if the operators are non-adjacent. Our method has proven to successfully find macro-operators of different lengths for six different benchmark domains. Also, our experiments highlighted the capital role of considering non-adjacent occurrences in the extraction of macro-operators.

Original language	English
Title of host publication	Knowledge Management and Acquisition for Intelligent Systems
Subtitle of host publication	17th Pacific Rim Knowledge Acquisition Workshop, PKAW 2020, Yokohama, Japan, January 7–8, 2021, Proceedings
Editors	Hiroshi Uehara, Takayasu Yamaguchi, Quan Bai
Publisher	Springer, Cham
Pages	30-45
Number of pages	16
Edition	1st
ISBN (Electronic)	9783030698867
ISBN (Print)	9783030698850
DOIs	https://doi.org/10.1007/978-3-030-69886-7_3
Publication status	Published - 20 Feb 2021
Externally published	Yes
Event	17th Pacific Rim Knowledge Acquisition Workshop, held in conjunction with the International Joint Conference on Artificial Intelligence - Pacific Rim International Conference on Artificial Intelligence - Yokohama, Japan Duration: 7 Jan 2021 → 8 Jan 2021 http://www.pkaw.org/pkaw2020/

Publication series

Name	Lecture Notes in Computer Science
Publisher	Springer
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	17th Pacific Rim Knowledge Acquisition Workshop, held in conjunction with the International Joint Conference on Artificial Intelligence - Pacific Rim International Conference on Artificial Intelligence
Abbreviated title	PKAW@IJCAI-PRICAI 2020
Country/Territory	Japan
City	Yokohama
Period	7/01/21 → 8/01/21
Internet address	http://www.pkaw.org/pkaw2020/

Access to Document

10.1007/978-3-030-69886-7_3Licence: Unspecified

Cite this

Castellanos-Paez, S., Rombourg, R., & Lalanda, P. (2021). ERA: Extracting Planning Macro-Operators from Adjacent and Non-adjacent Sequences. In H. Uehara, T. Yamaguchi, & Q. Bai (Eds.), Knowledge Management and Acquisition for Intelligent Systems: 17th Pacific Rim Knowledge Acquisition Workshop, PKAW 2020, Yokohama, Japan, January 7–8, 2021, Proceedings (1st ed., pp. 30-45). (Lecture Notes in Computer Science). Springer, Cham. https://doi.org/10.1007/978-3-030-69886-7_3

Castellanos-Paez, Sandra ; Rombourg, Romain ; Lalanda, Philippe. / ERA : Extracting Planning Macro-Operators from Adjacent and Non-adjacent Sequences. Knowledge Management and Acquisition for Intelligent Systems: 17th Pacific Rim Knowledge Acquisition Workshop, PKAW 2020, Yokohama, Japan, January 7–8, 2021, Proceedings. editor / Hiroshi Uehara ; Takayasu Yamaguchi ; Quan Bai. 1st. ed. Springer, Cham, 2021. pp. 30-45 (Lecture Notes in Computer Science).

@inproceedings{0aadb8257234455b88a61147d0d1a164,

title = "ERA: Extracting Planning Macro-Operators from Adjacent and Non-adjacent Sequences",

abstract = "Intuitively, Automated Planning systems capable of learning from previous experiences should be able to achieve better performance. One way to build on past experiences is to augment domains with macro-operators (i.e. frequent operator sequences). In most existing works, macros are generated from chunks of adjacent operators extracted from a set of plans. Although they provide some interesting results this type of analysis may provide incomplete results. In this paper, we propose ERA, an automatic extraction method for macro-operators from a set of solution plans. Our algorithm is domain and planner independent and can find all macro-operator occurrences even if the operators are non-adjacent. Our method has proven to successfully find macro-operators of different lengths for six different benchmark domains. Also, our experiments highlighted the capital role of considering non-adjacent occurrences in the extraction of macro-operators.",

keywords = "Automated planning, Macro-operators, Learning, Data mining",

author = "Sandra Castellanos-Paez and Romain Rombourg and Philippe Lalanda",

year = "2021",

month = feb,

day = "20",

doi = "10.1007/978-3-030-69886-7\_3",

language = "English",

isbn = "9783030698850",

series = "Lecture Notes in Computer Science",

publisher = "Springer, Cham",

pages = "30--45",

editor = "Uehara, \{Hiroshi \} and Takayasu Yamaguchi and Bai, \{Quan \}",

booktitle = "Knowledge Management and Acquisition for Intelligent Systems",

address = "Switzerland",

edition = "1st",

note = "17th Pacific Rim Knowledge Acquisition Workshop, held in conjunction with the International Joint Conference on Artificial Intelligence - Pacific Rim International Conference on Artificial Intelligence, PKAW@IJCAI-PRICAI 2020 ; Conference date: 07-01-2021 Through 08-01-2021",

url = "http://www.pkaw.org/pkaw2020/",

}

Castellanos-Paez, S, Rombourg, R & Lalanda, P 2021, ERA: Extracting Planning Macro-Operators from Adjacent and Non-adjacent Sequences. in H Uehara, T Yamaguchi & Q Bai (eds), Knowledge Management and Acquisition for Intelligent Systems: 17th Pacific Rim Knowledge Acquisition Workshop, PKAW 2020, Yokohama, Japan, January 7–8, 2021, Proceedings. 1st edn, Lecture Notes in Computer Science, Springer, Cham, pp. 30-45, 17th Pacific Rim Knowledge Acquisition Workshop, held in conjunction with the International Joint Conference on Artificial Intelligence - Pacific Rim International Conference on Artificial Intelligence, Yokohama, Japan, 7/01/21. https://doi.org/10.1007/978-3-030-69886-7_3

ERA: Extracting Planning Macro-Operators from Adjacent and Non-adjacent Sequences. / Castellanos-Paez, Sandra; Rombourg, Romain; Lalanda, Philippe.
Knowledge Management and Acquisition for Intelligent Systems: 17th Pacific Rim Knowledge Acquisition Workshop, PKAW 2020, Yokohama, Japan, January 7–8, 2021, Proceedings. ed. / Hiroshi Uehara; Takayasu Yamaguchi; Quan Bai. 1st. ed. Springer, Cham, 2021. p. 30-45 (Lecture Notes in Computer Science).

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

TY - GEN

T1 - ERA

T2 - 17th Pacific Rim Knowledge Acquisition Workshop, held in conjunction with the International Joint Conference on Artificial Intelligence - Pacific Rim International Conference on Artificial Intelligence

AU - Castellanos-Paez, Sandra

AU - Rombourg, Romain

AU - Lalanda, Philippe

PY - 2021/2/20

Y1 - 2021/2/20

N2 - Intuitively, Automated Planning systems capable of learning from previous experiences should be able to achieve better performance. One way to build on past experiences is to augment domains with macro-operators (i.e. frequent operator sequences). In most existing works, macros are generated from chunks of adjacent operators extracted from a set of plans. Although they provide some interesting results this type of analysis may provide incomplete results. In this paper, we propose ERA, an automatic extraction method for macro-operators from a set of solution plans. Our algorithm is domain and planner independent and can find all macro-operator occurrences even if the operators are non-adjacent. Our method has proven to successfully find macro-operators of different lengths for six different benchmark domains. Also, our experiments highlighted the capital role of considering non-adjacent occurrences in the extraction of macro-operators.

AB - Intuitively, Automated Planning systems capable of learning from previous experiences should be able to achieve better performance. One way to build on past experiences is to augment domains with macro-operators (i.e. frequent operator sequences). In most existing works, macros are generated from chunks of adjacent operators extracted from a set of plans. Although they provide some interesting results this type of analysis may provide incomplete results. In this paper, we propose ERA, an automatic extraction method for macro-operators from a set of solution plans. Our algorithm is domain and planner independent and can find all macro-operator occurrences even if the operators are non-adjacent. Our method has proven to successfully find macro-operators of different lengths for six different benchmark domains. Also, our experiments highlighted the capital role of considering non-adjacent occurrences in the extraction of macro-operators.

KW - Automated planning

KW - Macro-operators

KW - Learning

KW - Data mining

UR - https://www.scopus.com/pages/publications/85102780821

UR - https://link.springer.com/book/10.1007/978-3-030-69886-7

U2 - 10.1007/978-3-030-69886-7_3

DO - 10.1007/978-3-030-69886-7_3

M3 - Conference contribution

SN - 9783030698850

T3 - Lecture Notes in Computer Science

SP - 30

EP - 45

BT - Knowledge Management and Acquisition for Intelligent Systems

A2 - Uehara, Hiroshi

A2 - Yamaguchi, Takayasu

A2 - Bai, Quan

PB - Springer, Cham

Y2 - 7 January 2021 through 8 January 2021

ER -

Castellanos-Paez S, Rombourg R, Lalanda P. ERA: Extracting Planning Macro-Operators from Adjacent and Non-adjacent Sequences. In Uehara H, Yamaguchi T, Bai Q, editors, Knowledge Management and Acquisition for Intelligent Systems: 17th Pacific Rim Knowledge Acquisition Workshop, PKAW 2020, Yokohama, Japan, January 7–8, 2021, Proceedings. 1st ed. Springer, Cham. 2021. p. 30-45. (Lecture Notes in Computer Science). Epub 2021 Feb 19. doi: 10.1007/978-3-030-69886-7_3

ERA: Extracting Planning Macro-Operators from Adjacent and Non-adjacent Sequences

Abstract

Publication series

Conference

Access to Document

Other files and links

Fingerprint

Cite this