Introduction of Majority Vote of Neighborhood Conditions for Sneak Form Reinforcement Learning

摘要

Chain Form Reinforcement Learning (CFRL) was proposed for a reinforcement learning agent using low memory. In this paper, we introduce Sneak Form Reinforcement Learning (SFRL). SFRL is the method where we improve CFRL in terms of Contextual Learning. If a sequence of state-action pairs has a shortest path, a SFRL agent cuts and saves the path. To improve the performance of SFRL, we introduce Majority Vote of Neighborhood Conditions for SFRL and call this method MVNC. Majority Vote of Neighborhood Conditions is the rule which agent in an unknown state selects an action not at random but with circumjacent information. Our methods were made a comparison to Q-Learning and CFRL in several easy simulations. We examined performance and discussed thebest usage environment.