On the Evolution of Planner-Specific Macro Sets

Mauro Vallati, Lukas Chrpa, Ivan Serina

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

1 Citation (Scopus)

Abstract

In Automated Planning, generating macro-operators (macros) is a well-known reformulation approach that is used to speed-up the planning process. Most of the macro generation techniques aim for using the same set of generated macros on every problem instance of a given domain. This limits the usefulness of macros in scenarios where the environment and thus the structure of instances is dynamic, such as in real-world applications. Moreover, despite the wide availability of parallel processing units, there is a lack of approaches that can take advantage of multiple parallel cores, while exploiting macros.

In this paper we propose the Macro sets Evolution (MEvo) approach. MEvo has been designed for overcoming the aforementioned issues by exploiting multiple cores for combining promising macros –taken from a given pool– in different sets, while solving continuous streams of problem instances. Our empirical study, involving 5 state-of-the-art planning engines and a large number of planning instances, demonstrates the effectiveness of the proposed MEvo approach.
Original languageEnglish
Title of host publicationAI*AI 2017 Advances in Artificial Intelligence
Subtitle of host publicationXVIth International Conference of the Italian Association for Artificial Intelligence, Bari, Italy, November 14-17, 2017, Proceedings
EditorsFloriana Esposito, Roberto Basili, Stefano Ferilli, Francesca A. Lisi
PublisherSpringer Verlag
Pages443-454
Number of pages12
ISBN (Electronic)9783319701691
ISBN (Print)9783319701691
DOIs
Publication statusPublished - 7 Nov 2017
Event16th International Conference of the Italian Association for Artificial Intelligence - University of Bari, Bari, Italy
Duration: 14 Nov 201717 Nov 2017
Conference number: 16
http://aiia2017.di.uniba.it/ (Link to Conference Website)

Publication series

NameLecture Notes in Artificial Intelligence (LNAI)
PublisherSpringer
Volume10640
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference16th International Conference of the Italian Association for Artificial Intelligence
Abbreviated titleAI*IA 2017
CountryItaly
CityBari
Period14/11/1717/11/17
Internet address

Fingerprint

Macros
Planning
Availability
Engines
Processing

Cite this

Vallati, M., Chrpa, L., & Serina, I. (2017). On the Evolution of Planner-Specific Macro Sets. In F. Esposito, R. Basili, S. Ferilli, & F. A. Lisi (Eds.), AI*AI 2017 Advances in Artificial Intelligence: XVIth International Conference of the Italian Association for Artificial Intelligence, Bari, Italy, November 14-17, 2017, Proceedings (pp. 443-454). (Lecture Notes in Artificial Intelligence (LNAI); Vol. 10640). Springer Verlag. https://doi.org/10.1007/978-3-319-70169-1_33
Vallati, Mauro ; Chrpa, Lukas ; Serina, Ivan. / On the Evolution of Planner-Specific Macro Sets. AI*AI 2017 Advances in Artificial Intelligence: XVIth International Conference of the Italian Association for Artificial Intelligence, Bari, Italy, November 14-17, 2017, Proceedings. editor / Floriana Esposito ; Roberto Basili ; Stefano Ferilli ; Francesca A. Lisi. Springer Verlag, 2017. pp. 443-454 (Lecture Notes in Artificial Intelligence (LNAI)).
@inproceedings{7951a626d2b34199b64e9b275ff412fa,
title = "On the Evolution of Planner-Specific Macro Sets",
abstract = "In Automated Planning, generating macro-operators (macros) is a well-known reformulation approach that is used to speed-up the planning process. Most of the macro generation techniques aim for using the same set of generated macros on every problem instance of a given domain. This limits the usefulness of macros in scenarios where the environment and thus the structure of instances is dynamic, such as in real-world applications. Moreover, despite the wide availability of parallel processing units, there is a lack of approaches that can take advantage of multiple parallel cores, while exploiting macros.In this paper we propose the Macro sets Evolution (MEvo) approach. MEvo has been designed for overcoming the aforementioned issues by exploiting multiple cores for combining promising macros –taken from a given pool– in different sets, while solving continuous streams of problem instances. Our empirical study, involving 5 state-of-the-art planning engines and a large number of planning instances, demonstrates the effectiveness of the proposed MEvo approach.",
author = "Mauro Vallati and Lukas Chrpa and Ivan Serina",
year = "2017",
month = "11",
day = "7",
doi = "10.1007/978-3-319-70169-1_33",
language = "English",
isbn = "9783319701691",
series = "Lecture Notes in Artificial Intelligence (LNAI)",
publisher = "Springer Verlag",
pages = "443--454",
editor = "Floriana Esposito and Roberto Basili and Stefano Ferilli and Lisi, {Francesca A.}",
booktitle = "AI*AI 2017 Advances in Artificial Intelligence",

}

Vallati, M, Chrpa, L & Serina, I 2017, On the Evolution of Planner-Specific Macro Sets. in F Esposito, R Basili, S Ferilli & FA Lisi (eds), AI*AI 2017 Advances in Artificial Intelligence: XVIth International Conference of the Italian Association for Artificial Intelligence, Bari, Italy, November 14-17, 2017, Proceedings. Lecture Notes in Artificial Intelligence (LNAI), vol. 10640, Springer Verlag, pp. 443-454, 16th International Conference of the Italian Association for Artificial Intelligence, Bari, Italy, 14/11/17. https://doi.org/10.1007/978-3-319-70169-1_33

On the Evolution of Planner-Specific Macro Sets. / Vallati, Mauro; Chrpa, Lukas; Serina, Ivan.

AI*AI 2017 Advances in Artificial Intelligence: XVIth International Conference of the Italian Association for Artificial Intelligence, Bari, Italy, November 14-17, 2017, Proceedings. ed. / Floriana Esposito; Roberto Basili; Stefano Ferilli; Francesca A. Lisi. Springer Verlag, 2017. p. 443-454 (Lecture Notes in Artificial Intelligence (LNAI); Vol. 10640).

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

TY - GEN

T1 - On the Evolution of Planner-Specific Macro Sets

AU - Vallati, Mauro

AU - Chrpa, Lukas

AU - Serina, Ivan

PY - 2017/11/7

Y1 - 2017/11/7

N2 - In Automated Planning, generating macro-operators (macros) is a well-known reformulation approach that is used to speed-up the planning process. Most of the macro generation techniques aim for using the same set of generated macros on every problem instance of a given domain. This limits the usefulness of macros in scenarios where the environment and thus the structure of instances is dynamic, such as in real-world applications. Moreover, despite the wide availability of parallel processing units, there is a lack of approaches that can take advantage of multiple parallel cores, while exploiting macros.In this paper we propose the Macro sets Evolution (MEvo) approach. MEvo has been designed for overcoming the aforementioned issues by exploiting multiple cores for combining promising macros –taken from a given pool– in different sets, while solving continuous streams of problem instances. Our empirical study, involving 5 state-of-the-art planning engines and a large number of planning instances, demonstrates the effectiveness of the proposed MEvo approach.

AB - In Automated Planning, generating macro-operators (macros) is a well-known reformulation approach that is used to speed-up the planning process. Most of the macro generation techniques aim for using the same set of generated macros on every problem instance of a given domain. This limits the usefulness of macros in scenarios where the environment and thus the structure of instances is dynamic, such as in real-world applications. Moreover, despite the wide availability of parallel processing units, there is a lack of approaches that can take advantage of multiple parallel cores, while exploiting macros.In this paper we propose the Macro sets Evolution (MEvo) approach. MEvo has been designed for overcoming the aforementioned issues by exploiting multiple cores for combining promising macros –taken from a given pool– in different sets, while solving continuous streams of problem instances. Our empirical study, involving 5 state-of-the-art planning engines and a large number of planning instances, demonstrates the effectiveness of the proposed MEvo approach.

UR - http://www.springer.com/gp/book/9783319701684#aboutBook

U2 - 10.1007/978-3-319-70169-1_33

DO - 10.1007/978-3-319-70169-1_33

M3 - Conference contribution

SN - 9783319701691

T3 - Lecture Notes in Artificial Intelligence (LNAI)

SP - 443

EP - 454

BT - AI*AI 2017 Advances in Artificial Intelligence

A2 - Esposito, Floriana

A2 - Basili, Roberto

A2 - Ferilli, Stefano

A2 - Lisi, Francesca A.

PB - Springer Verlag

ER -

Vallati M, Chrpa L, Serina I. On the Evolution of Planner-Specific Macro Sets. In Esposito F, Basili R, Ferilli S, Lisi FA, editors, AI*AI 2017 Advances in Artificial Intelligence: XVIth International Conference of the Italian Association for Artificial Intelligence, Bari, Italy, November 14-17, 2017, Proceedings. Springer Verlag. 2017. p. 443-454. (Lecture Notes in Artificial Intelligence (LNAI)). https://doi.org/10.1007/978-3-319-70169-1_33