A Novel RL-Assisted Deep Learning Framework for Task-Informative Signals Selection and Classification for Spontaneous BCIs

In this article, we formulate the problem of estimating and selecting task-relevant temporal signal segments from a single electroencephalogram (EEG) trial in the form of a Markov decision process and propose a novel reinforcement-learning mechanism that can be combined with the existing deep-learning-based brain–computer interface methods. To be specific, we devise an actor–critic network such that an agent can determine which timepoints need to be used (informative) or discarded (uninformative) in composing the intention-related features in a given trial, and thus enhancing the intention identification performance. To validate the effectiveness of our proposed method, we conduct experiments with a publicly available big motor imagery (MI) dataset and apply our novel mechanism to various recent deep-learning architectures designed for MI classification. Based on the exhaustive experiments, we observe that our proposed method helped achieve statistically significant improvements in performance.