Study of emergence of strategies implemented by neural networksSupervisor: prof. RNDr. Jiří Pospíchal, DrSc.
This page shows details on the project. The primary projects are displayed together with a list of sub-projects.
|Project description:||The project from the area of distributed artificial intelligence will study emergence of strategies of multiagent systems, where single agents have their cognitive organ implemented by a neural network. Agents, placed in a given environment, must adapt to carry out prescribed activities (avoidance of obstacles, games, cooperation, ethnic conflicts, communication, and emergence of social structures). Neural network will be taught by reinforcement (reward/punishment) method. Activity of agents will be evaluated only after a longer time delay, when an external evaluator will be able to classify their behavior as successful or unsuccessful. The proposed method of adaptation will allow a systematic change of weight coefficients of neural networks in such a way, that the successfulness of their activities will progressively grow. The results of the research will be published in journals like e.g. The International Journal of Agent Technologies and Systems or The Journal of Artificial Societies and Social Simulation Scientific.|
|Kind of project:||VEGA ()|
|Department:||Institute of Applied Informatics (FIIT)|
|Project status:||Successfully completed|
|Project start date :||01. 01. 2010|
|Project close date:||31. 12. 2011|
|Number of workers in the project:||8|
|Number of official workers in the project:||0|