Complex Systems

A Distributed Learning Algorithm for Communication Development Download PDF

Edwin D. de Jong
Current address: Utrecht University, Decision Support Systems Group, Utrecht, The Netherlands. Electronic mail address: dejong@cs.uu.nl.

Luc Steels
Electronic mail address: steels@arti.vub.ac.be.
Vrije Universiteit Brussel,
Artificial Intelligence Laboratory,
Pleinlaan 2, B-1050 Brussels, Belgium

Abstract

We study the question of how a local learning algorithm, executed by multiple distributed agents, can lead to a global system of communication. First, the notion of a perfect communication system is defined. Next, two measures of communication system quality are specified. It is shown that maximization of these measures leads to perfect communication production. Based on this principle, local adaptation rules for communication development are constructed. The resulting stochastic algorithm is validated in computational experiments. Empirical analysis indicates that a mild degree of stochasticity is instrumental in reaching states that correspond to accurate communication.