Centralised and Decentralised Control of Video Game Agents

Samuel Robinson, Chloe M Barnes, Peter R. Lewis

Research output: Chapter in Book/Published conference outputConference publication


In this paper, the game of partially observable Ms. Pacman is used as a sandbox to evaluate Artificial Neural Networks (ANNs) that control multiple opponents (i.e. the ghosts). Comparisons between one central ANN that controls all ghosts, and multiple distinct ANNs, each controlling one ghost, are made. The NEAT algorithm is employed to evolve the ANNs. We find that chasing Ms. Pacman and exploring the map are both harder behaviours to learn for a centralised controller than for decentralised control. Further, both centralised and decentralised approaches produce vastly different behaviours for exploring the map. Novel techniques for comparing networks are also explored.
Original languageEnglish
Title of host publicationAdvances in Computational Intelligence Systems
Subtitle of host publicationEvolutionary Algorithms
Number of pages13
ISBN (Print)9783030870935, 9783030870942
Publication statusPublished - Jan 2022
Event20th UK Workshop on Computational Intelligence - Aberysytwyth, United Kingdom
Duration: 8 Sept 202110 Sept 2021

Publication series

NameAdvances in Intelligent Systems and Computing
ISSN (Print)2194-5357
ISSN (Electronic)2194-5365


Conference20th UK Workshop on Computational Intelligence
Country/TerritoryUnited Kingdom


Dive into the research topics of 'Centralised and Decentralised Control of Video Game Agents'. Together they form a unique fingerprint.

Cite this