Learning companion behaviors using reinforcement learning in games

Amir Ali Sharifi, Richard Zhao, Duane Szafron

Research output: Chapter in Book/Report/Conference proceedingConference contribution

5 Scopus citations

Abstract

Our goal is to enable Non Player Characters (NPC) in computer games to exhibit natural behaviors. The quality of behaviors affects the game experience especially in storybased games, which rely on player-NPC interactions. We used Reinforcement Learning to enable NPC companions to develop preferences for actions. We implemented our RL technique in BioWare Corp.'s Neverwinter Nights. Our experiments evaluate an NPC companion's behaviors regarding traps. Our method enables NPCs to rapidly learn reasonable behaviors and adapt to changes in the game.

Original languageEnglish (US)
Title of host publicationProceedings of the 6th AAAI Conference on Artificial Intelligence and Interactive Digital Entertainment, AIIDE 2010
Pages69-75
Number of pages7
StatePublished - Dec 1 2010
Event6th AAAI Conference on Artificial Intelligence and Interactive Digital Entertainment, AIIDE 2010 - Stanford, CA, United States
Duration: Oct 11 2010Oct 13 2010

Publication series

NameProceedings of the 6th AAAI Conference on Artificial Intelligence and Interactive Digital Entertainment, AIIDE 2010

Other

Other6th AAAI Conference on Artificial Intelligence and Interactive Digital Entertainment, AIIDE 2010
CountryUnited States
CityStanford, CA
Period10/11/1010/13/10

All Science Journal Classification (ASJC) codes

  • Artificial Intelligence
  • Visual Arts and Performing Arts

Fingerprint Dive into the research topics of 'Learning companion behaviors using reinforcement learning in games'. Together they form a unique fingerprint.

Cite this