发表机构
The Hong Kong University of Science and Technology（香港科技大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本文提出掩码频带功率预测（MBP）作为EEG自监督预训练目标，在睡眠分期任务上优于波形重建，验证了预训练目标应与下游任务物理量及时空尺度对齐。
AI 中文摘要
自监督EEG模型通常重建掩码波形或预测离散编码。我们研究了一种任务对齐的替代方案：掩码频带功率预测（MBP），该方案预测掩码通道-时间补丁的固定窄带对数谱能量。该目标保留了与睡眠分期相关的节律功率，同时避免了相位敏感的波形重建和学习型码本。在三个预训练种子下，我们使用匹配的骨干网络、预训练数据（2,388小时）和训练步骤比较了频带功率和波形目标，包括2x2分词器-目标设计。在ISRUC和HMC睡眠分期中，MBP在严格线性探针下，使用全部标签时超过原始和频带波形重建1.6-2.8个平衡准确率点，使用1%标签时超过4.7-7.3个点；目标效应超过分词器效应。其冻结特征达到0.7916/0.7425的平衡准确率，而匹配的丰富手工频谱基线为0.7636/0.7227，尽管在1%标签下差距约为一个点。完全微调达到0.8107/0.7669。这些增益并未扩展到所有具有频谱线索的任务，包括运动想象、抑郁筛查和警觉性回归。这些结果支持选择与下游标签相关的物理量以及空间和时间尺度相匹配的预训练目标。
英文摘要
Self-supervised EEG models often reconstruct masked waveforms or predict discrete codes. We study a task-aligned alternative: masked band-power prediction (MBP), which predicts fixed narrow-band log spectral energy for masked channel-time patches. This target retains rhythm power relevant to sleep staging while avoiding phase-sensitive waveform reconstruction and a learned codebook. Across three pretraining seeds, we compare band-power and waveform targets with matched backbones, pretraining data (2,388 hours), and training steps, including a 2x2 tokenizer-by-target design. On ISRUC and HMC sleep staging, MBP exceeds raw- and band-waveform reconstruction by 1.6-2.8 balanced-accuracy points with all labels and 4.7-7.3 points with 1% of labels under a strict linear probe; the target effect exceeds the tokenizer effect. Its frozen features reach 0.7916/0.7425 balanced accuracy, versus 0.7636/0.7227 for a matched rich handcrafted spectral baseline, although the gap is about one point with 1% of labels. Full fine-tuning reaches 0.8107/0.7669. The gains do not extend to every task with spectral cues, including motor imagery, depression screening, and vigilance regression. These results support choosing pretraining targets to match the physical quantities and spatial and temporal scales relevant to downstream labels.
Comments12 pages, 3 figures, 9 tables; includes an appendix
