Neural Networks Solving the Prisoner's Dilemma
Maciej Grzenda , Bohdan Macukow
AbstractPrisoner's Dilemma is an important model not only for game theory, but also psychology, economy and political science. Both to evolve successful strategies and simulate the behaviour of independent agents, the genetic algorithm has been used for a few years. In this paper the solution comprising the collaborative combination of neural network and genetic algorithm has been proposed. The evolutionary method for network learning has been selected to satisfy the requirements of the problem. Among other issues the ability of the algorithm to identify the input information indispensable for satisfying network performance has been tested.
|Journal series||Journal of Optoelectronics - Laser, ISSN 1005-0086|
|Publication size in sheets||0.3|
* presented citation count is obtained through Internet information analysis and it is close to the number calculated by the Publish or Perish system.