预处理器
计算机科学
人工智能
管道(软件)
可靠性(半导体)
机器学习
二进制数
鉴定(生物学)
数据挖掘
模式识别(心理学)
一致性(知识库)
灵敏度(控制系统)
二元分类
利用
矩阵分解
组分(热力学)
解码方法
反事实思维
特征提取
理论(学习稳定性)
成交(房地产)
数据预处理
独立成分分析
维数(图论)
集合(抽象数据类型)
边距(机器学习)
钥匙(锁)
作者
Dengzhe Hou,Zihao Wu,Lingyu Jiang,Zirui Li,Fangzhou Lin,Kazunori D. Yamada
摘要
Electroencephalography (EEG) is a cornerstone of brain-computer interfaces and clinical neuroscience, yet deep learning models are typically trained and evaluated under a single, unreported preprocessing pipeline. We formalize preprocessing choices as a counterfactual intervention space and show that EEG predictions are surprisingly unstable under this space: across six datasets spanning four paradigms, up to 42% of trial-level predictions flip when only the preprocessing changes, a variability that standard uncertainty methods do not explicitly quantify because they condition on a fixed preprocessing pipeline. We provide three tools to make this instability measurable, decomposable, and reducible. First, a Walsh-Hadamard decomposition of the 2^7 pipeline space reveals that sensitivity is near-additive in practice under the binary intervention design, enabling efficient step-by-step optimization. Second, we introduce Preprocessing Uncertainty (PU), a per-trial diagnostic that captures a dimension of instability complementary to model-based confidence. Third, we study Normalized Adaptive PGI (NA-PGI), a graph-structured regularizer that exploits the compositional structure of preprocessing interventions as one mitigation strategy with clear scope conditions.
科研通智能强力驱动
Strongly Powered by AbleSci AI