Search

Profit Sharing (PS), a reinforcement learning method that strongly reinforces successful experiences, has been shown to contribute to the improvement of learning speed when combined with a deep Q‐network (DQN). We expect a further improvement in learning speed by integrating PS‐based learning and Ape‐X DQN that has state‐of‐the‐art learning speed instead of the DQN. However, PS‐based learning does not use replay memory. In contrast, the Ape‐X DQN requires the use of replay memory because the exploration of the environment for collecting experiences and network training are performed asynchronously. In this study, we propose Learning‐accelerated Ape‐X, which integrates the Ape‐X DQN and PS‐based learning with some improvements including the use of replay memory. We show through numerical experiments that the proposed method improves the scores in Atari 2600 video games in a shorter time than the Ape‐X DQN. © 2020 Institute of Electrical Engineers of Japan. Published by John Wiley & Sons, Inc. [ABSTRACT FROM AUTHOR]

In this paper, the Rationally Oriented Forgettable Profit Sharing method (RFPS) for reinforcement learning is proposed. Although profit sharing (PS) provides good performances in real environments, its learning is often slow in long-term tasks because it is difficult to determine the appropriate discount rate which satisfies the Miyazaki rational theorem. There are several rationality-relaxed PS methods which work well for such tasks. However, these PS methods may result in many irrational loops. The proposed method fulfills rationality by forgetting the reinforced irrational loops. This method can be easily combined with ordinary PS methods and performs well in long-term tasks. Simulation results show that the proposed method can learn more efficiently than conventional PS methods. [ABSTRACT FROM AUTHOR]