Transfer learning action models by measuring the similarity of different domains

Hankui Zhuo; Qiang Yang; Li Lei

doi:10.1007/978-3-642-01307-2_70

Transfer learning action models by measuring the similarity of different domains

Hankui Zhuo^*, Qiang Yang, Li Lei

^*Corresponding author for this work

Department of Computer Science and Engineering

Research output: Chapter in Book/Conference Proceeding/Report › Conference Paper published in a book › peer-review

2 Citations (Scopus)

Abstract

AI planning requires action models to be given in advance. However, it is both time consuming and tedious for a human to encode the action models by hand using a formal language such as PDDL, as a result, learning action models is important for AI planning. On the other hand, the data being used to learn action models are often limited in planning domains, which makes the learning task very difficult. In this paper, we present a new algorithm to learn action models from plan traces by transferring useful information from other domains whose action models are already known. We present a method of building a metric to measure the shared information and transfer this information according to this metric. The larger the metric is, the bigger the information is transferred. In the experiment result, we show that our proposed algorithm is effective.

Original language	English
Title of host publication	13th Pacific-Asia Conference on Knowledge Discovery and Data Mining, PAKDD 2009
Pages	697-704
Number of pages	8
DOIs	https://doi.org/10.1007/978-3-642-01307-2_70
Publication status	Published - 2009
Event	13th Pacific-Asia Conference on Knowledge Discovery and Data Mining, PAKDD 2009 - Bangkok, Thailand Duration: 27 Apr 2009 → 30 Apr 2009

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	5476 LNAI
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	13th Pacific-Asia Conference on Knowledge Discovery and Data Mining, PAKDD 2009
Country/Territory	Thailand
City	Bangkok
Period	27/04/09 → 30/04/09

Access to Document

10.1007/978-3-642-01307-2_70

Cite this

Zhuo, H., Yang, Q., & Lei, L. (2009). Transfer learning action models by measuring the similarity of different domains. In 13th Pacific-Asia Conference on Knowledge Discovery and Data Mining, PAKDD 2009 (pp. 697-704). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 5476 LNAI). https://doi.org/10.1007/978-3-642-01307-2_70

@inproceedings{f46233aaadc1495c9a2732b56a0282b7,

title = "Transfer learning action models by measuring the similarity of different domains",

abstract = "AI planning requires action models to be given in advance. However, it is both time consuming and tedious for a human to encode the action models by hand using a formal language such as PDDL, as a result, learning action models is important for AI planning. On the other hand, the data being used to learn action models are often limited in planning domains, which makes the learning task very difficult. In this paper, we present a new algorithm to learn action models from plan traces by transferring useful information from other domains whose action models are already known. We present a method of building a metric to measure the shared information and transfer this information according to this metric. The larger the metric is, the bigger the information is transferred. In the experiment result, we show that our proposed algorithm is effective.",

author = "Hankui Zhuo and Qiang Yang and Li Lei",

year = "2009",

doi = "10.1007/978-3-642-01307-2\_70",

language = "English",

isbn = "3642013066",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

pages = "697--704",

booktitle = "13th Pacific-Asia Conference on Knowledge Discovery and Data Mining, PAKDD 2009",

note = "13th Pacific-Asia Conference on Knowledge Discovery and Data Mining, PAKDD 2009 ; Conference date: 27-04-2009 Through 30-04-2009",

}

Zhuo, H, Yang, Q & Lei, L 2009, Transfer learning action models by measuring the similarity of different domains. in 13th Pacific-Asia Conference on Knowledge Discovery and Data Mining, PAKDD 2009. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 5476 LNAI, pp. 697-704, 13th Pacific-Asia Conference on Knowledge Discovery and Data Mining, PAKDD 2009, Bangkok, Thailand, 27/04/09. https://doi.org/10.1007/978-3-642-01307-2_70

Transfer learning action models by measuring the similarity of different domains. / Zhuo, Hankui; Yang, Qiang; Lei, Li.
13th Pacific-Asia Conference on Knowledge Discovery and Data Mining, PAKDD 2009. 2009. p. 697-704 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 5476 LNAI).

Research output: Chapter in Book/Conference Proceeding/Report › Conference Paper published in a book › peer-review

TY - GEN

T1 - Transfer learning action models by measuring the similarity of different domains

AU - Zhuo, Hankui

AU - Yang, Qiang

AU - Lei, Li

PY - 2009

Y1 - 2009

N2 - AI planning requires action models to be given in advance. However, it is both time consuming and tedious for a human to encode the action models by hand using a formal language such as PDDL, as a result, learning action models is important for AI planning. On the other hand, the data being used to learn action models are often limited in planning domains, which makes the learning task very difficult. In this paper, we present a new algorithm to learn action models from plan traces by transferring useful information from other domains whose action models are already known. We present a method of building a metric to measure the shared information and transfer this information according to this metric. The larger the metric is, the bigger the information is transferred. In the experiment result, we show that our proposed algorithm is effective.

AB - AI planning requires action models to be given in advance. However, it is both time consuming and tedious for a human to encode the action models by hand using a formal language such as PDDL, as a result, learning action models is important for AI planning. On the other hand, the data being used to learn action models are often limited in planning domains, which makes the learning task very difficult. In this paper, we present a new algorithm to learn action models from plan traces by transferring useful information from other domains whose action models are already known. We present a method of building a metric to measure the shared information and transfer this information according to this metric. The larger the metric is, the bigger the information is transferred. In the experiment result, we show that our proposed algorithm is effective.

UR - https://www.scopus.com/pages/publications/67650683377

U2 - 10.1007/978-3-642-01307-2_70

DO - 10.1007/978-3-642-01307-2_70

M3 - Conference Paper published in a book

AN - SCOPUS:67650683377

SN - 3642013066

SN - 9783642013065

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 697

EP - 704

BT - 13th Pacific-Asia Conference on Knowledge Discovery and Data Mining, PAKDD 2009

T2 - 13th Pacific-Asia Conference on Knowledge Discovery and Data Mining, PAKDD 2009

Y2 - 27 April 2009 through 30 April 2009

ER -

Zhuo H, Yang Q, Lei L. Transfer learning action models by measuring the similarity of different domains. In 13th Pacific-Asia Conference on Knowledge Discovery and Data Mining, PAKDD 2009. 2009. p. 697-704. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). doi: 10.1007/978-3-642-01307-2_70

Transfer learning action models by measuring the similarity of different domains

Abstract

Publication series

Conference

Access to Document

Other files and links

Fingerprint

Cite this