A Python-Based MPI Framework for Exploring an Adaptive Fuzzy-Agent Approach to Simulating Large-Scale Non-Cooperative Games

In this article, we describe how to construct a large scale simulation system using the standard message passing interface (MPI) framework which can effectively explore the simulated players' strategy search spaces (i.e., to identify "good" strategies within particular "games&quo...

Full description

Saved in:
Bibliographic Details
Main Authors: Millman, E., Budakoglu, C., Neville, S.W.
Format: Conference Proceeding
Language:English
Subjects:
Computational and artificial intelligence
Computational modeling
Computer security
Computer simulation
Ecosystems
Fuzzy logic
Game theory
Genetic algorithms
Large-scale systems
Process control
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:In this article, we describe how to construct a large scale simulation system using the standard message passing interface (MPI) framework which can effectively explore the simulated players' strategy search spaces (i.e., to identify "good" strategies within particular "games" out of large sets of potential strategies) using genetic algorithms. We demonstrate how to create "intelligent" players who are capable of adapting their behaviors as the game evolves, given the problematic nature of identifying "good" strategies a priori using fuzzy logic. We prove these two concepts by building a scalable predator and prey simulation framework.
ISSN:0840-7789
2576-7046
DOI:10.1109/CCECE.2007.348