A Distributed Learning Algorithm for Communication Development Export

@misc{citeulike:524171, 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.}, author = {de Jong, Edwin D. and Steels, Luc}, citeulike-article-id = {524171}, citeulike-linkout-0 = {http://citeseer.ist.psu.edu/701286.html}, citeulike-linkout-1 = {http://citeseer.lcs.mit.edu/701286.html}, citeulike-linkout-2 = {http://citeseer.ifi.unizh.ch/701286.html}, citeulike-linkout-3 = {http://citeseer.comp.nus.edu.sg/701286.html}, keywords = {language\_game, lexicon}, posted-at = {2006-02-28 13:09:39}, priority = {3}, title = {A Distributed Learning Algorithm for Communication Development}, url = {http://citeseer.ist.psu.edu/701286.html} }

TY - GEN ID - citeulike:524171 L3 - citeulike-article-id:524171 N2 - 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. TI - A Distributed Learning Algorithm for Communication Development KW - language_game KW - lexicon AU - de Jong, Edwin AU - Steels, Luc UR - http://citeseer.ist.psu.edu/701286.html ER -