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

Automated translation of dynamic pro...

Mauch, Holger.

Linked to FindBook

Google Book

Amazon

博客來

Automated translation of dynamic programming problems to JAVA code and their solution via an intermediate Petri net representation.

Record Type:	Electronic resources : Monograph/item
Title/Author:	Automated translation of dynamic programming problems to JAVA code and their solution via an intermediate Petri net representation./
Author:	Mauch, Holger.
Description:	202 p.
Notes:	Source: Dissertation Abstracts International, Volume: 66-04, Section: B, page: 2160.
Contained By:	Dissertation Abstracts International66-04B.
Subject:	Computer Science. -
Online resource:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3171060
ISBN:	0542077302

Automated translation of dynamic programming problems to JAVA code and their solution via an intermediate Petri net representation.
Mauch, Holger.

Automated translation of dynamic programming problems to JAVA code and their solution via an intermediate Petri net representation. - 202 p.

Source: Dissertation Abstracts International, Volume: 66-04, Section: B, page: 2160.

Thesis (Ph.D.)--University of Hawai'i at Manoa, 2005.

Dynamic programming (DP) is a very general technique for solving optimization problems. In this dissertation, we develop a software system called "DP2PN2Solver" that automates many of the tasks a user encounters in order to solve an instance of a discrete optimization problem by means of DP. The user gives a specification of the DP functional equation in a special purpose high-level language. Then the specification is transformed into an intermediate Petri net (PN) representation for the discrete DP problem to be solved. After consistency checks the PN is translated into JAVA code (or some other object code that can be run in an additional step, e.g. a spreadsheet file), which solves the problem instance. Theoretical investigations examine the space of DP problems amenable to automation, as well as theory about the specialized intermediate PN representation called a "Bellman net". Bellman nets are defined in two flavors; one is based upon the high-level or colored PN definition, the other is based upon the low-level place/transition net definition. These two approaches are then put in contrast. It is shown that the novel low-level definition is sufficient for many common discrete DP problems, while the high-level definition is general enough to handle an extremely broad class of discrete DP problems.

ISBN: 0542077302Subjects--Topical Terms:

626642
Computer Science.

Automated translation of dynamic programming problems to JAVA code and their solution via an intermediate Petri net representation.
LDR:02257nmm 2200277 4500 001 1812825
005 20060427132636.5
008 130610s2005 eng d
020 $a 0542077302
035 $a (UnM)AAI3171060
035 $a AAI3171060
040 $a UnM $c UnM
100 1 $a Mauch, Holger. $3 939126
245 1 0 $a Automated translation of dynamic programming problems to JAVA code and their solution via an intermediate Petri net representation.
300 $a 202 p.
500 $a Source: Dissertation Abstracts International, Volume: 66-04, Section: B, page: 2160.
500 $a Chairperson: Art Lew.
502 $a Thesis (Ph.D.)--University of Hawai'i at Manoa, 2005.
520 $a Dynamic programming (DP) is a very general technique for solving optimization problems. In this dissertation, we develop a software system called "DP2PN2Solver" that automates many of the tasks a user encounters in order to solve an instance of a discrete optimization problem by means of DP. The user gives a specification of the DP functional equation in a special purpose high-level language. Then the specification is transformed into an intermediate Petri net (PN) representation for the discrete DP problem to be solved. After consistency checks the PN is translated into JAVA code (or some other object code that can be run in an additional step, e.g. a spreadsheet file), which solves the problem instance. Theoretical investigations examine the space of DP problems amenable to automation, as well as theory about the specialized intermediate PN representation called a "Bellman net". Bellman nets are defined in two flavors; one is based upon the high-level or colored PN definition, the other is based upon the low-level place/transition net definition. These two approaches are then put in contrast. It is shown that the novel low-level definition is sufficient for many common discrete DP problems, while the high-level definition is general enough to handle an extremely broad class of discrete DP problems.
590 $a School code: 0085.
650 4 $a Computer Science. $3 626642
650 4 $a Operations Research. $3 626629
690 $a 0984
690 $a 0796
710 2 0 $a University of Hawai'i at Manoa. $3 1017511
773 0 $t Dissertation Abstracts International $g 66-04B.
790 1 0 $a Lew, Art, $e advisor
790 $a 0085
791 $a Ph.D.
792 $a 2005
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3171060