arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 1903.11673cs.LGeess.SPstat.ML
收藏

EEG生物识别中的对抗性深度学习

Adversarial Deep Learning in EEG Biometrics

Northeastern University(东北大学);Mitsubishi Electric Research Laboratories(三菱电机研究实验室)

Ozan Ozdenizci, Ye Wang, Toshiaki Koike-Akino, Deniz Erdogmus

摘要:

用于基于脑电图(EEG)大脑活动的人员识别的深度学习方法遇到利用EEG内的时间相关结构或记录会话特定变化的问题。此外,最近的方法大多基于单次会话EEG数据进行训练和评估。我们从不变表示学习的角度来解决这个问题。我们提出了一种对抗性推理方法来扩展这种深度学习模型,以学习会话不变的个人区分表示,可以在纵向可用性方面提供鲁棒性。在深度卷积网络中使用对抗性学习,我们根据纵向收集的EEG数据对我们的方法进行了经验评估和改进,用于从半秒EEG时期进行人员识别。

英文摘要:

Deep learning methods for person identification based on electroencephalographic (EEG) brain activity encounters the problem of exploiting the temporally correlated structures or recording session specific variability within EEG. Furthermore, recent methods have mostly trained and evaluated based on single session EEG data. We address this problem from an invariant representation learning perspective. We propose an adversarial inference approach to extend such deep learning models to learn session-invariant person-discriminative representations that can provide robustness in terms of longitudinal usability. Using adversarial learning within a deep convolutional network, we empirically assess and show improvements with our approach based on longitudinally collected EEG data for person identification from half-second EEG epochs.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑