uncertainty conditions

Game model of self-organizing of multiagent systems

The game model of multi-agent systems of self-organizing in the conditions of uncertainty is developed. The formulation of a stochastic game problem is carried out, criteria of self-organizing of strategies of players are defined, a recurrent method, algorithm and software of learning of multi-agent system to simulate the synchronised rhythmic luminescence of a colony of fireflies are developed.

Модель стохастичної гри нейроагентів

The neuroagent game model of collective decision-making in the conditions of uncertainty is developed. The formulation of stochastic game is executed. Adaptive learning methods of artificial neural networks without the teacher are used for the game solving. The convergence of neuroagent stochastic game is confirmed by results of computer experiment. Influences of parameters of game model on the neuroagent learning rate are investigated.

Ігровий метод синхронізації подій в мультиагентних системах

The adaptive game method of events synchronization in multiagent systems in the conditions of uncertainty is developed. The essence of a method consists in alignment of delays of approach of events on the basis of supervision of actions of the next players. The formulation of stochastic game is executed and game algorithm for its solving is developed. Influences of parameters on convergence of a game method are investigated by means of computer experiment.

Matrix stochastic game with Q-learning

The model of matrix stochastic game for decision-making in the conditions of uncertainty is developed. The method of Q-learning for stochastic game solving with a priori unknown gains matrices is offered. The formulation of a game problem is executed. The Markovian recurrent method and algorithm for the game solving are described. Results of computer modelling of stochastic game with Q-learning are received and analysed.

Neuroagent model of decision-making

The neuroagent model of decision-making in the conditions of uncertainty is investigated. Adaptive methods of an artificial neural network learning without the teacher are considered. The algorithm and program model of neuroagent decision-making are developed. Efficiency of neuroagent decision-making has been confirmed by results of computer experiment. Influences of parameters of model on the neuroagent learning rate are investigated.