应用平稳矩阵logistic回归优化脑机接口单次EEG辨识

In addition to the noisy and limited spatial resolution characteristics of the EEG signal, the intrinsic non-stationarity in the EEG data makes the single-trial EEG classi?cation an even more challenging problem in brain-computer interface (BCI). Variations of the signal properties within a session often result in deteriorated classi?cation performance. This is mainly attributed to the reason that the routine feature extraction or classi?cation method does not take the changes in the signal into account. Although several extensions to the standard feature extraction method have been proposed to reduce the sensitivity to non-stationarity in data, they optimizes different objective functions from that of the subsequent classi?cation model, thereby the extracted features may not be optimized for the classi?cation. In this paper, we propose an approach that directly optimizes the classi?er's discriminativity and robustness against non-stationarity in the EEG data with a single optimization paradigm, and show that it can greatly improve the performance, in particular for the subjects who have dif?culty in controlling a BCI. Moreover, the experimental results on two benchmark data sets demonstrate that our approach signi?cantly outperforms the state-of-the-art approaches in reducing classi?cation error rates.