Making Optimal Decisions with Learning Method Based on Fuzzy Logic

This work aims to develop the functions and structure of intelligent agents with fuzzy logic based on methods of reinforcement learning and their application to solve practical problems of decision-making in conditions of uncertainty. To achieve this goal, it is used fuzzy inference based on production rules knowledge bases to develop an algorithm for the operation of an intelligent agent. The empirical verification of the developed algorithm efficacy is carried out for the task of agent finding the optimal value of the function in the cell space with the determination of the direction of movement according to the rules of fuzzy logic.