Replacing the Reinforcement Learning (RL) to the Auto Reinforcement Learning (AutoRL) Algorithms to Find the Optimal Structure of Business Processes in the Bank

Abstract

The paper provides a description of the algorithm allowing the shift from RL to AutoRL in the search for the optimal path in a banking business process grid graph. The article discusses which attributes of the environment are universal and which are not. It describes how it is necessary to change the Reward so that it becomes universal for any business process in the bank. #COMESYSO1120.