Domain Model Acquisition with Missing Information and Noisy Data

Peter Gregory, Alan Lindsay, Julie Porteous

Research output: Contribution to conferencePaperResearchpeer-review

6 Downloads (Pure)

Abstract

In this work, we address the problem of learning planning domain models from example action traces that contain missing and noisy data. In many situations, the action traces that form the input to domain model acquisition systems are sourced from observations or even natural language descriptions of plans. It is often the case that these observations are noisy and incomplete. Therefore, making domain model acquisition systems that are robust to such data is crucial. Previous approaches to this problem have relied upon having access to the underlying state in the input plans. We lift this assumption and provide a system that does not require any state information. We build upon the LOCM family of algorithms, which also lift this assumption in the deterministic version of the domain model acquisition problem, to provide a domain model acquisition system that learns domain models from noisy plans with missing information.
Original languageEnglish
Pages-
Publication statusPublished - 18 Jun 2017
EventWorkshop on Knowledge Engineering for Planning and Scheduling (KEPS). The 27th International Conference on Automated Planning and Scheduling (ICAPS) - Pittsburgh, United States
Duration: 18 Jun 201723 Jun 2017

Conference

ConferenceWorkshop on Knowledge Engineering for Planning and Scheduling (KEPS). The 27th International Conference on Automated Planning and Scheduling (ICAPS)
Abbreviated titleKEPS 2017
CountryUnited States
CityPittsburgh
Period18/06/1723/06/17

Fingerprint

Planning

Bibliographical note

Author can archive post-print, The paper should not be posted until it is formally published with AAAI. http://www.aaai.org/ojs/index.php/aimagazine/about/editorialPolicies#a

Cite this

Gregory, P., Lindsay, A., & Porteous, J. (2017). Domain Model Acquisition with Missing Information and Noisy Data. -. Paper presented at Workshop on Knowledge Engineering for Planning and Scheduling (KEPS). The 27th International Conference on Automated Planning and Scheduling (ICAPS), Pittsburgh, United States.
Gregory, Peter ; Lindsay, Alan ; Porteous, Julie. / Domain Model Acquisition with Missing Information and Noisy Data. Paper presented at Workshop on Knowledge Engineering for Planning and Scheduling (KEPS). The 27th International Conference on Automated Planning and Scheduling (ICAPS), Pittsburgh, United States.
@conference{05e9cf2344f74f8499d051d99bb22488,
title = "Domain Model Acquisition with Missing Information and Noisy Data",
abstract = "In this work, we address the problem of learning planning domain models from example action traces that contain missing and noisy data. In many situations, the action traces that form the input to domain model acquisition systems are sourced from observations or even natural language descriptions of plans. It is often the case that these observations are noisy and incomplete. Therefore, making domain model acquisition systems that are robust to such data is crucial. Previous approaches to this problem have relied upon having access to the underlying state in the input plans. We lift this assumption and provide a system that does not require any state information. We build upon the LOCM family of algorithms, which also lift this assumption in the deterministic version of the domain model acquisition problem, to provide a domain model acquisition system that learns domain models from noisy plans with missing information.",
author = "Peter Gregory and Alan Lindsay and Julie Porteous",
note = "Author can archive post-print, The paper should not be posted until it is formally published with AAAI. http://www.aaai.org/ojs/index.php/aimagazine/about/editorialPolicies#a; Workshop on Knowledge Engineering for Planning and Scheduling (KEPS). The 27th International Conference on Automated Planning and Scheduling (ICAPS), KEPS 2017 ; Conference date: 18-06-2017 Through 23-06-2017",
year = "2017",
month = "6",
day = "18",
language = "English",
pages = "--",

}

Gregory, P, Lindsay, A & Porteous, J 2017, 'Domain Model Acquisition with Missing Information and Noisy Data' Paper presented at Workshop on Knowledge Engineering for Planning and Scheduling (KEPS). The 27th International Conference on Automated Planning and Scheduling (ICAPS), Pittsburgh, United States, 18/06/17 - 23/06/17, pp. -.

Domain Model Acquisition with Missing Information and Noisy Data. / Gregory, Peter; Lindsay, Alan; Porteous, Julie.

2017. - Paper presented at Workshop on Knowledge Engineering for Planning and Scheduling (KEPS). The 27th International Conference on Automated Planning and Scheduling (ICAPS), Pittsburgh, United States.

Research output: Contribution to conferencePaperResearchpeer-review

TY - CONF

T1 - Domain Model Acquisition with Missing Information and Noisy Data

AU - Gregory, Peter

AU - Lindsay, Alan

AU - Porteous, Julie

N1 - Author can archive post-print, The paper should not be posted until it is formally published with AAAI. http://www.aaai.org/ojs/index.php/aimagazine/about/editorialPolicies#a

PY - 2017/6/18

Y1 - 2017/6/18

N2 - In this work, we address the problem of learning planning domain models from example action traces that contain missing and noisy data. In many situations, the action traces that form the input to domain model acquisition systems are sourced from observations or even natural language descriptions of plans. It is often the case that these observations are noisy and incomplete. Therefore, making domain model acquisition systems that are robust to such data is crucial. Previous approaches to this problem have relied upon having access to the underlying state in the input plans. We lift this assumption and provide a system that does not require any state information. We build upon the LOCM family of algorithms, which also lift this assumption in the deterministic version of the domain model acquisition problem, to provide a domain model acquisition system that learns domain models from noisy plans with missing information.

AB - In this work, we address the problem of learning planning domain models from example action traces that contain missing and noisy data. In many situations, the action traces that form the input to domain model acquisition systems are sourced from observations or even natural language descriptions of plans. It is often the case that these observations are noisy and incomplete. Therefore, making domain model acquisition systems that are robust to such data is crucial. Previous approaches to this problem have relied upon having access to the underlying state in the input plans. We lift this assumption and provide a system that does not require any state information. We build upon the LOCM family of algorithms, which also lift this assumption in the deterministic version of the domain model acquisition problem, to provide a domain model acquisition system that learns domain models from noisy plans with missing information.

M3 - Paper

SP - -

ER -

Gregory P, Lindsay A, Porteous J. Domain Model Acquisition with Missing Information and Noisy Data. 2017. Paper presented at Workshop on Knowledge Engineering for Planning and Scheduling (KEPS). The 27th International Conference on Automated Planning and Scheduling (ICAPS), Pittsburgh, United States.