传感器故障下的泛化能力：Tokenization + Transformers 构建更鲁棒的潜在空间
Generalizability Under Sensor Failure: Tokenization + Transformers Enable More Robust Latent Spaces
AI总结：
针对神经数据在传感器故障下的泛化挑战，研究收集脑电图数据集并对比 EEGNet 与 TOTEM 模型，发现基于 tokenization 和 transformer 的 TOTEM 在所有泛化情形下表现更优或持平，证实分词机制有助于构建鲁棒的潜在空间。
摘要：
神经科学的一个主要目标是发现可泛化的神经数据表征。这一目标受到记录会话（如环境）、受试者（如不同的神经结构）和传感器（如传感器噪声）等变异性的挑战。近期研究开始解决跨会话和受试者的泛化问题，但鲜少研究对在神经科学实验中极为普遍的传感器故障的鲁棒性。为解决这些泛化维度，我们首先收集了包含大量会话、受试者和传感器的脑电图数据集，随后研究了两种时间序列模型：EEGNet 和 TOTEM。EEGNet 是广泛使用的卷积神经网络，而 TOTEM 是离散时间序列分词器和 transformer 模型。我们发现 TOTEM 在所有泛化情形下均优于或匹敌 EEGNet。最后，通过对 TOTEM 潜在码本的分析，我们观察到 tokenization 能够实现泛化。
英文摘要：
A major goal in neuroscience is to discover neural data representations that generalize. This goal is challenged by variability along recording sessions (e.g. environment), subjects (e.g. varying neural structures), and sensors (e.g. sensor noise), among others. Recent work has begun to address generalization across sessions and subjects, but few study robustness to sensor failure which is highly prevalent in neuroscience experiments. In order to address these generalizability dimensions we first collect our own electroencephalography dataset with numerous sessions, subjects, and sensors, then study two time series models: EEGNet (Lawhern et al., 2018) and TOTEM (Talukder et al., 2024). EEGNet is a widely used convolutional neural network, while TOTEM is a discrete time series tokenizer and transformer model. We find that TOTEM outperforms or matches EEGNet across all generalizability cases. Finally through analysis of TOTEM's latent codebook we observe that tokenization enables generalization.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
