A spectral audit framework reveals task-dependent aperiodic reliance across EEG and ECG deep learning
频谱审计框架揭示EEG和ECG深度学习中任务依赖的非周期性依赖
机构 * Indian Knowledge Systems and Mental Health Applications (IKSMHA) Center, Indian Institute of Technology Mandi(印度理工学院曼迪分校印度知识体系与心理健康应用中心) ; School of Computing and Electrical Engineering, Indian Institute of Technology Mandi(印度理工学院曼迪分校计算与电气工程学院)
专题命中 健康监测 :pathology(abstract);分类 cs.LG、eess.SP
AI总结 提出频谱审计框架,结合非周期/周期分解、相位保持傅里叶干预等,发现深度学习模型对非周期成分的依赖是任务依赖且架构通用的,在睡眠-觉醒分类中影响显著,临床异常检测中中等,运动想象中最小,并扩展到ECG。
Comments 25 pages, being prepared for submission to peer-reviewed journal