Adapting Brain Signals With Reinforcement Learning Strategies for Brain Computer Interfaces