Abstract

There is increasing awareness in the planning community that depending on complete models impedes the applicability of planning technology in many real world domains where the burden of specifying complete domain models is too high. In this paper, we consider the problem of generating robust and accurate plans, when the agent only has access to incomplete domain models, supplanted by a set of successful plan cases. We will develop two classes of approaches – one case-based and the other model-based. ML-CBP is a case-based approach that leverages the incomplete model and the plan cases to solve a new problem directly by affecting case-level transfer. RIM is a model-based approach that uses the incomplete model and the plan cases to first learn a more complete model. This model contains both primitive actions as well as macro-operators that are derived from the plan cases. The learned model is then used in conjunction with an off-the-shelf planner to solve new problems. We present a comprehensive evaluation of the two approaches, both to characterize their relative tradeoffs, and to quantify their advances over existing approaches.

Original languageEnglish (US)
Pages (from-to)1-21
Number of pages21
JournalArtificial Intelligence
Volume246
DOIs
StatePublished - May 1 2017

Fingerprint

planning model
Planning
Reaction injection molding
planning
Macros

Keywords

  • Action model learning
  • AI planning
  • Case-based planning

ASJC Scopus subject areas

  • Language and Linguistics
  • Linguistics and Language
  • Artificial Intelligence

Cite this

Model-lite planning : Case-based vs. model-based approaches. / Zhuo, Hankz Hankui; Kambhampati, Subbarao.

In: Artificial Intelligence, Vol. 246, 01.05.2017, p. 1-21.

Research output: Contribution to journalArticle

@article{5e27d1effb90478c84e878f3c303e147,
title = "Model-lite planning: Case-based vs. model-based approaches",
abstract = "There is increasing awareness in the planning community that depending on complete models impedes the applicability of planning technology in many real world domains where the burden of specifying complete domain models is too high. In this paper, we consider the problem of generating robust and accurate plans, when the agent only has access to incomplete domain models, supplanted by a set of successful plan cases. We will develop two classes of approaches – one case-based and the other model-based. ML-CBP is a case-based approach that leverages the incomplete model and the plan cases to solve a new problem directly by affecting case-level transfer. RIM is a model-based approach that uses the incomplete model and the plan cases to first learn a more complete model. This model contains both primitive actions as well as macro-operators that are derived from the plan cases. The learned model is then used in conjunction with an off-the-shelf planner to solve new problems. We present a comprehensive evaluation of the two approaches, both to characterize their relative tradeoffs, and to quantify their advances over existing approaches.",
keywords = "Action model learning, AI planning, Case-based planning",
author = "Zhuo, {Hankz Hankui} and Subbarao Kambhampati",
year = "2017",
month = "5",
day = "1",
doi = "10.1016/j.artint.2017.01.004",
language = "English (US)",
volume = "246",
pages = "1--21",
journal = "Artificial Intelligence",
issn = "0004-3702",
publisher = "Elsevier",

}

TY - JOUR

T1 - Model-lite planning

T2 - Case-based vs. model-based approaches

AU - Zhuo, Hankz Hankui

AU - Kambhampati, Subbarao

PY - 2017/5/1

Y1 - 2017/5/1

N2 - There is increasing awareness in the planning community that depending on complete models impedes the applicability of planning technology in many real world domains where the burden of specifying complete domain models is too high. In this paper, we consider the problem of generating robust and accurate plans, when the agent only has access to incomplete domain models, supplanted by a set of successful plan cases. We will develop two classes of approaches – one case-based and the other model-based. ML-CBP is a case-based approach that leverages the incomplete model and the plan cases to solve a new problem directly by affecting case-level transfer. RIM is a model-based approach that uses the incomplete model and the plan cases to first learn a more complete model. This model contains both primitive actions as well as macro-operators that are derived from the plan cases. The learned model is then used in conjunction with an off-the-shelf planner to solve new problems. We present a comprehensive evaluation of the two approaches, both to characterize their relative tradeoffs, and to quantify their advances over existing approaches.

AB - There is increasing awareness in the planning community that depending on complete models impedes the applicability of planning technology in many real world domains where the burden of specifying complete domain models is too high. In this paper, we consider the problem of generating robust and accurate plans, when the agent only has access to incomplete domain models, supplanted by a set of successful plan cases. We will develop two classes of approaches – one case-based and the other model-based. ML-CBP is a case-based approach that leverages the incomplete model and the plan cases to solve a new problem directly by affecting case-level transfer. RIM is a model-based approach that uses the incomplete model and the plan cases to first learn a more complete model. This model contains both primitive actions as well as macro-operators that are derived from the plan cases. The learned model is then used in conjunction with an off-the-shelf planner to solve new problems. We present a comprehensive evaluation of the two approaches, both to characterize their relative tradeoffs, and to quantify their advances over existing approaches.

KW - Action model learning

KW - AI planning

KW - Case-based planning

UR - http://www.scopus.com/inward/record.url?scp=85012066318&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85012066318&partnerID=8YFLogxK

U2 - 10.1016/j.artint.2017.01.004

DO - 10.1016/j.artint.2017.01.004

M3 - Article

AN - SCOPUS:85012066318

VL - 246

SP - 1

EP - 21

JO - Artificial Intelligence

JF - Artificial Intelligence

SN - 0004-3702

ER -