The Evolution of Foraging in an Open-Ended Simulation Environment
Authors
Abstract
Throughout the last decades, Darwin?s theory of natural se- lection has fueled a vast amount of research in the field of computer science, and more specifically in artificial intelligence. The majority of this work has focussed on artificial selection, rather than on natural se- lection. In this paper we study the evolution of agents? controllers in an open-ended scenario. To that end, we set up a multi-agent simulation inspired by the ant foraging task, and evolve the agents? brain (a rule list) without any explicit fitness function. We show that the agents do evolve sustainable foraging behaviors in this environment, and discuss some evolutionary conditions that seem to be important to achieve these results.
Keywords
open-ended evolution, artificial life, complex systems, multi-agent systems
Subject
Artificial Life
Conference
Progress in Artificial Intelligence: 15th Portuguese Conference on Artificial Intelligence, October 2011
Cited by
No citations found