Centralised and Decentralised Control of Video Game Agents

被引：0

作者：

Robinson, Sam G. ^{[1
]}

Barnes, Chloe M. ^{[1
]}

Lewis, Peter R. ^{[2
]}

机构：

[1] Aston Univ, Birmingham B4 7ET, W Midlands, England

[2] Ontario Tech Univ, Oshawa, ON L1G 0C5, Canada

来源：

ADVANCES IN COMPUTATIONAL INTELLIGENCE SYSTEMS | 2022年 / 1409卷

关键词：

Neuroevolution; Game playing; Network comparisons;

D O I：

10.1007/978-3-030-87094-2_10

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

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.

引用

页码：108 / 120

页数：13