ProtoGIB-Workload：跨被试学习特定工作负载的神经拓扑原型
ProtoGIB-Workload: Learning Workload-Specific Neural Topology Prototypes across Subjects
AI总结：
本文针对EEG工作负载识别中跨被试泛化差的问题，提出ProtoGIB-Workload框架，结合SGIB和CTS实现稳定的神经拓扑原型学习，在多数据集LOSO实验中提升跨被试Macro-F1分数平均5.15%。
AI中文摘要：
可靠的基于脑电图（EEG）的心理工作负载识别对自适应以人为中心的系统至关重要，但实际部署要求模型能够泛化到训练期间未见过的用户。尽管功能连接图被广泛用于捕捉与工作负载相关的神经交互，但其本质上会将任务相关结构与被试特定的生理特征及样本级噪声纠缠在一起，这种纠缠常导致模型学习到结构捷径，严重降低跨被试泛化能力。为解决该问题，本文提出ProtoGIB-Workload，一种新型框架，用于明确正则化和对齐图结构以实现与被试无关的工作负载识别。该方法引入随机图信息瓶颈（SGIB），将密集相关性先验压缩为紧凑的、任务相关的子图，过滤与输入相关的冗余；关键的是，为防止保留被试特定的虚假边，本文提出类条件拓扑稳定器（CTS），利用EEG数据的固定电极坐标，CTS直接作用于图生成概率，鼓励具有相同工作负载类别的不同被试间的边生成统计量保持一致。在两个公开EEG工作负载数据集和一个内部的空中交通管制员EEG认知负载数据集上，采用严格的留一被试（LOSO）协议进行的大量实验表明，ProtoGIB-Workload显著优于最先进的基于时间和图的基线，将跨被试Macro-F1分数平均提高5.15%（最高达6.34%），进一步分析证实该方法成功提取了稳定的、跨被试一致的神经连接模式。
英文摘要：
Reliable electroencephalography (EEG)-based mental workload recognition is crucial for adaptive human-centered systems, yet practical deployment requires models to generalize to users unseen during training. Although functional connectivity graphs are widely adopted to capture workload-related neural interactions, they inherently entangle task-relevant structures with subject-specific physiological traits and sample-level noise. This entanglement often leads models to learn structural shortcuts, severely degrading cross-subject generalization. To address this, we propose ProtoGIB-Workload, a novel framework that explicitly regularizes and aligns graph structures for subject-independent workload recognition. Our approach introduces a Stochastic Graph Information Bottleneck (SGIB) to compress dense correlation priors into compact, task-relevant subgraphs, filtering out input-related redundancy. Crucially, to prevent the retention of subject-specific spurious edges, we propose a Class-Conditional Topology Stabilizer (CTS). Leveraging the fixed electrode coordinates of EEG data, CTS operates directly on graph-generation probabilities to encourage consistent edge-generation statistics across different subjects sharing the same workload class. Extensive experiments on two public EEG workload datasets and one in-house EEG cognitive load dataset of air traffic controllers under strict leave-one-subject-out (LOSO) protocols demonstrate that ProtoGIB-Workload significantly outperforms state-of-the-art temporal and graph-based baselines, improving the cross-subject Macro-F1 score by an average of 5.15% (up to 6.34%). Further analyses confirm that our method successfully extracts stable, cross-subject consistent neural connectivity patterns.
