Högskolan i Skövde

his.sePublications
Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • apa-cv
  • 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
AI Training Grounds: A study of adaptive AI in RPGs with low stochastic dependence
University of Skövde, School of Informatics.
2019 (English)Independent thesis Basic level (degree of Bachelor), 20 credits / 30 HE creditsStudent thesis
Abstract [en]

Artificial intelligence is used in video games for controlling AI agents but often follow a static pre-defined behaviour. An AI capable of online learning would be able to adapt their behaviour to whatever they may encounter. ‘Dynamic scripting’ is one such technique designed to be used in games with random outcomes. This study seeks to examine if there is a measurable difference in the learning efficiency of dynamic scripting between games with stochastic outcomes and those with deterministic outcomes. A CRPG was made with dynamic scripting implemented in charge of a team of AI agents and a toggle to force deterministic or stochastic game rules. A series of encounters against a static AI was then performed until the dynamic team was deemed effective enough and the learning efficiency measured by how long this took. The results suggest a difference favouring dynamic scripting in deterministic environments.

Place, publisher, year, edition, pages
2019. , p. 18
Keywords [en]
video games, artificial intelligence, reinforcement learning, dynamic scripting
National Category
Computer Engineering
Identifiers
URN: urn:nbn:se:his:diva-17008OAI: oai:DiVA.org:his-17008DiVA, id: diva2:1322486
Subject / course
Informationsteknologi
Educational program
Computer Game Development - Design
Examiners
Available from: 2019-06-19 Created: 2019-06-10 Last updated: 2023-08-31Bibliographically approved

Open Access in DiVA

No full text in DiVA

Search in DiVA

By author/editor
Jensen, Aksel
By organisation
School of Informatics
Computer Engineering

Search outside of DiVA

GoogleGoogle Scholar

urn-nbn

Altmetric score

urn-nbn
Total: 220 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • apa-cv
  • 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