Reinforcement Learning-Driven Feedback Optimisation Algorithms for Brain-Computer Interfaces

Brain-computer interface (BCI) systems have important application value in the field of neurorehabilitation and intelligent control, but there is still room for optimisation of the existing feedback mechanism in terms of classification accuracy, information transfer rate (ITR) and user adaptability. This study proposes a feedback optimisation algorithm based on reinforcement learning to construct an adaptive feedback adjustment mechanism with PPO as the core to improve the interaction efficiency of BCI system. The experiments are validated based on BCI Competition IV Dataset 2a, and the results show that the method can improve the classification accuracy to 87.3%, the ITR to 31.5 bits/min, and the user response time to 1.28s, which are significantly improved compared with the traditional methods. The study shows that the feedback strategy optimised by reinforcement learning can effectively improve the performance of the BCI system and provide new technical support for the future application of BCI in the field of medical rehabilitation and intelligent interaction.