[2603.26713] Boundary-aware Prototype-driven Adversarial Alignment for Cross-Corpus EEG Emotion Recognition

Computer Science > Machine Learning arXiv:2603.26713 (cs) [Submitted on 18 Mar 2026] Title:Boundary-aware Prototype-driven Adversarial Alignment for Cross-Corpus EEG Emotion Recognition Authors:Guangli Li, Canbiao Wu, Na Tian, Li Zhang, Zhen Liang View a PDF of the paper titled Boundary-aware Prototype-driven Adversarial Alignment for Cross-Corpus EEG Emotion Recognition, by Guangli Li and 4 other authors View PDF HTML (experimental) Abstract:Electroencephalography (EEG)-based emotion recognition suffers from severe performance degradation when models are transferred across heterogeneous datasets due to physiological variability, experimental paradigm differences, and device inconsistencies. Existing domain adversarial methods primarily enforce global marginal alignment and often overlook class-conditional mismatch and decision boundary distortion, limiting cross-corpus generalization. In this work, we propose a unified Prototype-driven Adversarial Alignment (PAA) framework for cross-corpus EEG emotion recognition. The framework is progressively instantiated in three configurations: PAA-L, which performs prototype-guided local class-conditional alignment; PAA-C, which further incorporates contrastive semantic regularization to enhance intra-class compactness and inter-class separability; and PAA-M, the full boundary-aware configuration that integrates dual relation-aware classifiers within a three-stage adversarial optimization scheme to explicitly refine controversial sam...