EEGMoE:一个领域脱的专家混合模型,用于自我监督的EEG表示学习
IEEE transactions on neural networks and learning systems
|January 19, 2026
概括
我们介绍了EEG专家混合 (EEGMoE),这是一个新的自我监督模型,它从电脑电图 (EEG) 数据中学习了共享和域特定的表示. 这种方法提高了各种任务的概括性,例如情绪识别和运动图像.
科学领域:
- 神经科学是一个神经科学.
- 人工智能的人工智能
- 机器学习 机器学习
背景情况:
- 电脑电图 (EEG) 分析的深度学习模型通常是特定任务的,这限制了它们的概括性.
- 当前的预训练方法统一数据格式,但忽视了EEG表示中的关键领域特定细节.
- 大规模的EEG数据包括不同的领域,需要采用捕捉共同和独特特征的方法.
研究的目的:
- 开发一个自我监督的预训练模型,用于EEG表示学习,将特定领域的信息分离开来.
- 从不同的EEG数据集中学习域共享和域特定的表示.
- 提高EEG模型在各种任务和数据集中的通用性.
主要方法:
- 提出EEG专家组合 (EEGMoE),这是一个基于变压器的域脱编码器,使用专家组合 (MoE) 块.
- 实施针对特定专家组的Top-K路由以及在教育部区块内的共享专家组的软路由.
- 在各种EEG数据集上预训练EEGMoE,并在情绪识别,运动图像分类和心理工作负载检测任务上进行微调/验证.
主要成果:
- 在三个公共数据集上,EEGMoE的性能优于最新的模型,用于情绪识别,运动图像和心理工作负载检测.
- 显示强大的泛化能力,以新的,未见的EEG领域.
- 广泛的实验和可视化证实了解散域特定表示的有效性.
结论:
- EEGMoE成功地学习了共享和特定域的EEG表示,从而提高了模型的通用性.
- 拟议的域脱方法对于充分利用大规模多域EEG数据的全部潜力至关重要.
- 这些发现强调了解表示对于各种应用中的强大的EEG分析的重要性.
相关概念视频
¹³C NMR: ¹H–¹³C Decoupling
1.8K
The probability of having two carbon-13 atoms next to each other is negligible because of the low natural abundance of carbon-13. Consequently, peak splitting due to carbon-carbon spin-spin coupling is not observed in spectra. However, protons up to three sigma bonds away split the carbon signal according to the n+1 rule, resulting in complicated spectra.
A broadband decoupling technique is used to simplify these complex, sometimes overlapping, signals. Broadband decoupling relies on a...
A broadband decoupling technique is used to simplify these complex, sometimes overlapping, signals. Broadband decoupling relies on a...
1.8K
Mixtures of Acids
21.5K
The pH of a solution containing an acid can be determined using its acid dissociation constant and its initial concentration. If a solution contains two different acids, then its pH can be determined using one of several methods depending upon the relative strength of the acids and their dissociation constants.
A Mixture of a Strong Acid and a Weak Acid
In a mixture of a strong acid and a weak acid, the strong acid dissociates completely and becomes a source of almost all the hydronium ions...
A Mixture of a Strong Acid and a Weak Acid
In a mixture of a strong acid and a weak acid, the strong acid dissociates completely and becomes a source of almost all the hydronium ions...
21.5K
Mixtures of Acids
1.1K
The pH of a solution containing an acid can be determined using its acid dissociation constant and initial concentration. If a solution contains two different acids, then its pH can be determined using one of several methods depending on the relative strength of the acids and their dissociation constants.
In a strong and weak acid mixture, the strong acid dissociates completely and becomes a source of almost all the hydronium ions present in the solution. In contrast, the weak acid shows...
In a strong and weak acid mixture, the strong acid dissociates completely and becomes a source of almost all the hydronium ions present in the solution. In contrast, the weak acid shows...
1.1K
Control Volume and System Representations
1.5K
Two key frameworks are employed to analyze mass, energy, and momentum transfer: the control volume approach and the system approach. These frameworks offer different perspectives, depending on whether the focus is on a specific region in space (control volume approach) or a defined mass of fluid (system approach).
The control volume approach considers a stationary region in space through which fluid flows. This region is bounded by a control surface. For instance, in the case of water...
The control volume approach considers a stationary region in space through which fluid flows. This region is bounded by a control surface. For instance, in the case of water...
1.5K
Fast Decoupled and DC Powerflow
734
The fast decoupled power flow method addresses contingencies in power system operations, such as generator outages or transmission line failures. This method provides quick power flow solutions, essential for real-time system adjustments. Fast decoupled power flow algorithms simplify the Jacobian matrix by neglecting certain elements, leading to two sets of decoupled equations:
734
State Space Representation
536
The frequency-domain technique, commonly used in analyzing and designing feedback control systems, is effective for linear, time-invariant systems. However, it falls short when dealing with nonlinear, time-varying, and multiple-input multiple-output systems. The time-domain or state-space approach addresses these limitations by utilizing state variables to construct simultaneous, first-order differential equations, known as state equations, for an nth-order system.
Consider an RLC circuit, a...
Consider an RLC circuit, a...
536


