MAGE:面向可泛化多时间尺度脑电表示的基础模型
原标题:MAGE: Towards Generalizable Multi-timescale EEG Representations
AI 摘要
研究团队提出 MAGE(Multi-timescale Adaptive Generalization for EEG),一个面向脑电信号的多时间尺度基础模型。它通过多时间尺度分词、多时间尺度注意力机制和可训练聚合模块,动态捕捉并融合不同时间尺度的信息,以克服现有 EEG 基础模型依赖固定时间窗口的局限。在七个数据集上的实验显示,MAGE 以更少参数持续优于现有 EEG 基础模型,并支持不同下游任务受益于不同时间尺度的假设。
正文节选
MAGE: Towards Generalizable Multi-timescale EEG Representations Abstract Electroencephalography (EEG) data plays an important role in understanding human brain activity, with various major applications. While current approaches, which largely build on EEG foundation models, have demonstrated strong performance across datasets and tasks, they typically rely on a fixed temporal scale. This design choice limits their ability to capture characteristics of neural dynamic behavior, often spanning acro