How Much Capacity Does EEG Denoising Need? Ultra-Compact Networks reveal Benchmark Saturation and Metric-Utility Gap
脑电图去噪需要多少容量?超紧凑网络揭示基准饱和与度量-效用差距
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 EEG解码 :EEG(title,summary_cn);BCI(summary_cn,abstract);分类 eess.SP、cs.LG
AI总结 通过固定架构仅改变通道宽度(1.05K-40.26K参数),发现EEG去噪重建性能在3-6.5K参数时饱和,且重建度量不预测下游BCI效用,超紧凑模型(33-46KB)适用于边缘部署。
Comments 17 pages, will be submitted to peer-reviewed journal