Towards Evolving Cooperative Behavior with Neural Controllers
نویسندگان
چکیده
A self-organizing system achieves a global system behavior via local interactions between its entities without centralized control [1]. In other words, the entities, or agents, have to cooperate in order to achieve a global result. The problem of cooperation can be observed in different domains and on many levels, especially abundant in nature. It essentially boils down to the conflict of interest between a group of individuals and between the individual itself. In social studies and experimental economics, public goods game proved to be a standard approach to investigate cooperation in various conditions and parameters[2][3]. The basic idea of this game is that a group of subjects starting with a given number of tokens secretly choose how many of their private tokens to put into the public pot. Each subject keeps the remaining tokens plus an even split of the tokens in the pot. Usually, the pot undergoes some function (e.g., multiplied by with a factor) before it gets redistributed to encourage cooperation. In the simplest instance of this problem the following behavior can be observed: the participants start with cooperation but in the following rounds the overall donated money tends to zero. This phenomena has been extensively studied in literature proposing new methods like sanctioning between the members of a group when antisocial behavior is observed [4]. In this paper we examine a different version of the public goods game providing us a suitable testbed to evolve neural controllers as players while monitoring their willingness for cooperation.
منابع مشابه
Evolving an Integrated Phototaxis and Hole-avoidance Behavior for a Swarm-bot
This article is on the subject of evolving neural network controllers for cooperative, mobile robots. We evolve controllers for combined hole-avoidance and phototaxis in a group of physically connected, autonomous robots called s-bots, each with limited sensing capabilities. We take a systematic approach to finding a suitable fitness function, an appropriate neural network structure, and we exp...
متن کاملEvolving Swimming Controllers for a Simulated Lamprey with Inspiration from Neurobiology Accepted for Publication in Adaptive Behavior
This paper presents how neural swimming controllers for a simulated lamprey can be developed using evolutionary algorithms. A genetic algorithm is used for evolving the architecture of a connectionist model which determines the muscular activity of a simulated body in interaction with water. This work is inspired by the biological model developed by Ekeberg which reproduces the central pattern ...
متن کاملEvolving Neural Network Controllers for a Team of Self-Organizing Robots
Self-organizing systems obtain a global system behavior via typically simple local interactions among a number of components or agents, respectively. The emergent service often displays properties like adaptability, robustness, and scalability, which makes the self-organizing paradigm interesting for technical applications like cooperative autonomous robots. The behavior for the local interacti...
متن کاملEvolution of RF-Signal Cognition for Wheeled Mobile Robots using Pareto Multi-objective Optimization
This article describes a simulation model in which a multi-objective approach is utilized for evolving an artificial neural networks (ANNs) controller for an autonomous mobile robot. A mobile robot is simulated in a 3D, physics-based environment for the RF-localization behavior. The elitist Pareto-frontier Differential Evolution (PDE) algorithm is used to generate the Pareto optimal set of ANNs...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
عنوان ژورنال:
دوره شماره
صفحات -
تاریخ انتشار 2009