Fundamental Study of Collaborative Learning Consisting of Reinforcement Learning and Brain Signal in BCI

Abstract

Reinforcement learning has been studied as an unsupervised learning method. Alternatively, BCI(Brain Computer Interface) comes into the research limelight. However, nonsynchronous spontaneous action potentials and evoked action potentials exist contain brain signal, and we need an interface model which exists between brain and machine for control and stability. In this paper, we propose a collaborative learning system consisting of reinforcement learning and brain signal in BCI. Brain signal is interpreted as a deliberate assignment of the subject, and we utilize reinforcement learning in control and stability for BCI. We apply the collaborative learning to maze problem and show the usefulness of the proposed system.