发表机构
University of Washington（华盛顿大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
研究对比贝叶斯完全池化模型与频率主义基线用于跨受试者运动想象脑电图分类，经多指标分析发现前者可靠性有统计学改善但实际不显著，计算成本高，表明其实际益处有限，跨受试者和会话部分池化是更有前景方向。
AI 中文摘要
脑机接口长期以来一直寻求免校准操作，但分类器通常仅通过辨别能力进行基准测试，而忽略了预测概率是否校准良好。鉴于非平稳脑电图信号以及分布变化下过度自信的点估计分类器的风险，这是一个有意义的差距。我们进行了一项大规模研究，将贝叶斯完全池化模型与频率主义基线进行对比，用于跨20个数据集的跨受试者、左手与右手运动想象脑电图分类。六个频率主义管道分别与一个类似的贝叶斯管道配对，共享相同的特征工程，通过马尔可夫链蒙特卡罗后验采样进行拟合。主要指标是布里尔分数，分解为可靠性和分辨率，同时还有用于辨别的AUROC和用于清晰度的香农熵。每个指标通过随机效应荟萃分析进行分析，并通过留一法影响分析进行验证。贝叶斯完全池化在可靠性方面有统计学上但实际上不显著的改善，预测不确定性增加（清晰度降低）；布里尔分数、分辨率和辨别能力没有显著差异。所有指标的研究间异质性较低，不过可靠性结果对留一法去除敏感。我们还分析了计算成本，发现贝叶斯管道消耗的能量大约是其频率主义对应物的13倍，相对于普通家用电器来说，这个成本仍然适中。这些结果表明，仅贝叶斯完全池化在跨受试者运动想象分类中提供的实际益处有限，跨受试者和会话进行部分池化是未来工作更有前景的方向。
英文摘要
Brain-computer interfaces (BCIs) have long sought calibration-free operation, yet classifiers are typically benchmarked on discrimination alone. Discrimination is blind to calibration, a meaningful gap given that electroencephalogram (EEG) signals are nonstationary and point-estimate classifiers can become overconfident under distribution shift. We conducted a large-scale study contrasting Bayesian complete-pooling models against frequentist baselines for cross-subject, left-hand versus right-hand motor imagery EEG classification across 20 datasets. Each of six frequentist pipelines was paired with an analogous Bayesian pipeline sharing identical feature engineering, fit via Markov chain Monte Carlo posterior sampling. Our primary metric was the Brier score, decomposed into reliability and resolution, alongside the area under the receiver operating characteristic curve for discrimination and Shannon entropy for sharpness. For each metric we fit a random-effects meta-analysis with Knapp-Hartung adjustment, verified by leave-one-out influence analysis. Bayesian complete-pooling produced statistically significant improvements in reliability and increases in predictive uncertainty (lower sharpness), but 95% confidence intervals bounded both effects near zero, and neither held significance when datasets with flagged posterior sampling were excluded. Brier score, resolution, and discrimination showed no significant differences. Between-study heterogeneity was low across all metrics. Bayesian pipelines consumed roughly thirteen times more energy than their frequentist counterparts, a cost that remains modest relative to common household appliances. Bayesian complete-pooling alone offers limited benefit for cross-subject motor imagery classification, but its modest cost makes partial-pooling across subjects and sessions a feasible next step.
