kth.sePublications
Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
A Formally Motivated Algorithm for Synthesis of Strategies in Multi-Agent Games of Imperfect Information Against Nature
KTH, School of Electrical Engineering and Computer Science (EECS).
KTH, School of Electrical Engineering and Computer Science (EECS).
2023 (English)Independent thesis Basic level (degree of Bachelor), 10 credits / 15 HE creditsStudent thesisAlternative title
En formellt motiverad algoritm för syntes av strategier för fler-agent-spel med okomplett information mot miljön (Swedish)
Abstract [en]

In this project we study games where multiple agents cooperate to reach some common team objective. The agents do not have perfect information about the game state, and they cannot communicate during the course of the game. In this report, we have limited the scope to simple objectives where the goal is to reach any member of a specified set of game states. To achieve this, we translate the multi-agent game of imperfect information to individual games of perfect information for each agent using the method Multi-Agent Knowledge-Based Subset Contruction (MKBSC). In these games, each representing the view of an individual agent, we find all possibly-winning strategies, and then combine them into a surely-winning multi-agent strategy profile for the original game, if one exists. Unlike previous work, we also provide a formal motivation for the correctness of this method.

Abstract [sv]

I det här projektet studerar vi spel där flera agenter samarbetar för att nå ett visst gemensamt mål. De har inte perfekt information om spelets tillstånd, och de kan inte kommunicera under spelets gång. I denna rapport har vi begränsat omfattningen till enkla mål, där målet är att nå ett av en viss uppsättning tillstånd. För att uppnå detta översätter vi fleragentspelet med ofullständig information till individuella spel med perfekt information med hjälp av metoden Multi-Agent Knowledge-Based Subset Construction"(MKBSC) för att översätta ett fleragentspel med ofullständig information till individuella spel med perfekt information för varje agent. I dessa spel, som vart och ett representerar perspektivet hos en individuell agent, hittar vi alla potentiellt-vinnande strategier och kombinerar dem sedan till en definitivt-vinnande strategi i det ursprungliga spelet, om en sådan existerar. Till skillnad från tidigare arbeten tillhandahåller vi också en formell motivering till att denna metod är korrekt.

Place, publisher, year, edition, pages
2023. , p. 34
Series
TRITA-EECS-EX ; 2023:277
National Category
Computer and Information Sciences
Identifiers
URN: urn:nbn:se:kth:diva-330765OAI: oai:DiVA.org:kth-330765DiVA, id: diva2:1778364
Supervisors
Examiners
Available from: 2023-07-27 Created: 2023-07-01 Last updated: 2023-07-27Bibliographically approved

Open Access in DiVA

fulltext(744 kB)111 downloads
File information
File name FULLTEXT01.pdfFile size 744 kBChecksum SHA-512
f2204583fb39ca2ebebd0b3bc80f326788917295985f3c0684f30419a1a01fae9caab816efcaca5a1e6d00b1fdb7d818d4ed4db1d1c52aa02b2c7e529b0edfae
Type fulltextMimetype application/pdf

By organisation
School of Electrical Engineering and Computer Science (EECS)
Computer and Information Sciences

Search outside of DiVA

GoogleGoogle Scholar
Total: 111 downloads
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

urn-nbn

Altmetric score

urn-nbn
Total: 6515 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf