東華大學圖書館 |

Language: English

Help

回圖書館首頁

手機版館藏查詢

Back

Switch To: Labeled | MARC Mode | ISBD

Reasoning and Learning with Probabil...

Wang, Yi.

Linked to FindBook

Google Book

Amazon

博客來

Reasoning and Learning with Probabilistic Answer Set Programming.

Record Type:	Electronic resources : Monograph/item
Title/Author:	Reasoning and Learning with Probabilistic Answer Set Programming./
Author:	Wang, Yi.
Published:	Ann Arbor : ProQuest Dissertations & Theses, : 2019,
Description:	263 p.
Notes:	Source: Dissertations Abstracts International, Volume: 80-11, Section: B.
Contained By:	Dissertations Abstracts International80-11B.
Subject:	Computer science. -
Online resource:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=13862749
ISBN:	9781392140376

Reasoning and Learning with Probabilistic Answer Set Programming.
Wang, Yi.

Reasoning and Learning with Probabilistic Answer Set Programming. - Ann Arbor : ProQuest Dissertations & Theses, 2019 - 263 p.

Source: Dissertations Abstracts International, Volume: 80-11, Section: B.

Thesis (Ph.D.)--Arizona State University, 2019.

This item must not be sold to any third party vendors.

Knowledge Representation (KR) is one of the prominent approaches to Artificial Intelligence (AI) that is concerned with representing knowledge in a form that computer systems can utilize to solve complex problems. Answer Set Programming (ASP), based on the stable model semantics, is a widely-used KR framework that facilitates elegant and efficient representations for many problem domains that require complex reasoning.However, while ASP is effective on deterministic problem domains, it is not suitable for applications involving quantitative uncertainty, for example, those that require probabilistic reasoning. Furthermore, it is hard to utilize information that can be statistically induced from data with ASP problem modeling.This dissertation presents the language LPMLN, which is a probabilistic extension of the stable model semantics with the concept of weighted rules, inspired by Markov Logic. An LPMLN program defines a probability distribution over "soft" stable models, which may not satisfy all rules, but the more rules with the bigger weights they satisfy, the bigger their probabilities. LPMLN takes advantage of both ASP and Markov Logic in a single framework, allowing representation of problems that require both logical and probabilistic reasoning in an intuitive and elaboration tolerant way.This dissertation establishes formal relations between LPMLN and several other formalisms, discusses inference and weight learning algorithms under LPMLN, and presents systems implementing the algorithms. LPMLN systems can be used to compute other languages translatable into LPMLN.The advantage of LPMLN for probabilistic reasoning is illustrated by a probabilistic extension of the action language BC+, called pBC+, defined as a high-level notation of LPMLN for describing transition systems. Various probabilistic reasoning about transition systems, especially probabilistic diagnosis, can be modeled in pBC+ and computed using LPMLN systems. pBC+ is further extended with the notion of utility, through a decision-theoretic extension of LPMLN, and related with Markov Decision Process (MDP) in terms of policy optimization problems. pBC+ can be used to represent (PO)MDP in a succinct and elaboration tolerant way, which enables planning with (PO)MDP algorithms in action domains whose description requires rich KR constructs, such as recursive definitions and indirect effects of actions.

ISBN: 9781392140376Subjects--Topical Terms:

523869
Computer science.

Reasoning and Learning with Probabilistic Answer Set Programming.
LDR:03444nmm a2200313 4500 001 2264258
005 20200423112931.5
008 220629s2019 ||||||||||||||||| ||eng d
020 $a 9781392140376
035 $a (MiAaPQ)AAI13862749
035 $a (MiAaPQ)asu:18866
035 $a AAI13862749
040 $a MiAaPQ $c MiAaPQ
100 1 $a Wang, Yi. $3 1035436
245 1 0 $a Reasoning and Learning with Probabilistic Answer Set Programming.
260 1 $a Ann Arbor : $b ProQuest Dissertations & Theses, $c 2019
300 $a 263 p.
500 $a Source: Dissertations Abstracts International, Volume: 80-11, Section: B.
500 $a Publisher info.: Dissertation/Thesis.
500 $a Advisor: Lee, Joohyung.
502 $a Thesis (Ph.D.)--Arizona State University, 2019.
506 $a This item must not be sold to any third party vendors.
520 $a Knowledge Representation (KR) is one of the prominent approaches to Artificial Intelligence (AI) that is concerned with representing knowledge in a form that computer systems can utilize to solve complex problems. Answer Set Programming (ASP), based on the stable model semantics, is a widely-used KR framework that facilitates elegant and efficient representations for many problem domains that require complex reasoning.However, while ASP is effective on deterministic problem domains, it is not suitable for applications involving quantitative uncertainty, for example, those that require probabilistic reasoning. Furthermore, it is hard to utilize information that can be statistically induced from data with ASP problem modeling.This dissertation presents the language LPMLN, which is a probabilistic extension of the stable model semantics with the concept of weighted rules, inspired by Markov Logic. An LPMLN program defines a probability distribution over "soft" stable models, which may not satisfy all rules, but the more rules with the bigger weights they satisfy, the bigger their probabilities. LPMLN takes advantage of both ASP and Markov Logic in a single framework, allowing representation of problems that require both logical and probabilistic reasoning in an intuitive and elaboration tolerant way.This dissertation establishes formal relations between LPMLN and several other formalisms, discusses inference and weight learning algorithms under LPMLN, and presents systems implementing the algorithms. LPMLN systems can be used to compute other languages translatable into LPMLN.The advantage of LPMLN for probabilistic reasoning is illustrated by a probabilistic extension of the action language BC+, called pBC+, defined as a high-level notation of LPMLN for describing transition systems. Various probabilistic reasoning about transition systems, especially probabilistic diagnosis, can be modeled in pBC+ and computed using LPMLN systems. pBC+ is further extended with the notion of utility, through a decision-theoretic extension of LPMLN, and related with Markov Decision Process (MDP) in terms of policy optimization problems. pBC+ can be used to represent (PO)MDP in a succinct and elaboration tolerant way, which enables planning with (PO)MDP algorithms in action domains whose description requires rich KR constructs, such as recursive definitions and indirect effects of actions.
590 $a School code: 0010.
650 4 $a Computer science. $3 523869
690 $a 0984
710 2 $a Arizona State University. $b Computer Science. $3 1676136
773 0 $t Dissertations Abstracts International $g 80-11B.
790 $a 0010
791 $a Ph.D.
792 $a 2019
793 $a English
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=13862749