AI 中文总结
本文提出结合神经常微分方程的多粒度流形对比学习模型MGMCL，在三个公开脑电情感数据集上实现跨主体情感识别的最优性能，较现有方法有显著提升。
AI 中文摘要
跨主体脑电图（EEG）情感识别因存在显著的个体间差异，且现有方法采用离散式表述而忽略情感的连续性，仍具挑战性。现有方法均在欧氏空间中操作，聚焦于边缘分布对齐，无法保留跨主体的情感语义结构。本文提出MGMCL，将情感识别重新概念化为在对称正定（SPD）黎曼流形上学习连续表示。该框架在实例、情感和轨迹三个层级引入多粒度流形对比学习，同时保留语义顺序；利用流形上的神经常微分方程建模连续情感动态；采用格罗莫夫-瓦瑟斯坦（Gromov-Wasserstein）流形对齐实现跨主体泛化；通过弱监督学习从离散标签中实现连续的效价-唤醒-优势度预测。在三个公开数据集上开展的大量实验表明，该方法达到了当前最优性能：在SEED数据集上准确率为91.23%，SEED-IV数据集上为73.82%，DEAP数据集上为76.38%，分别较之前的最优方法实现了1.89%、1.66%和1.28%的一致提升。
英文摘要
Cross-subject electroencephalogram (EEG)-based emotion recognition remains challenging due to substantial inter-individual variability and discrete formulation that overlooks affective continuity. Existing methods operate in Euclidean space and focus on marginal distribution alignment, failing to preserve the semantic structure of emotions across subjects. This article proposes MGMCL, reconceptualizing emotion recognition as learning continuous representations on symmetric positive definite (SPD) Riemannian manifolds. The frame?work introduces multi-granularity manifold contrastive learning at instance, emotion, and trajectory levels while preserving semantic ordering. Neural ordinary differential equations on manifolds model continuous emotion dynamics. Cross-subject generalization employs Gromov-Wasserstein manifold alignment. Weakly-supervised learning enables continuous valence-arousal-dominance prediction from discrete labels. Extensive experiments on three public datasets demonstrate state-of-the-art performance: 91.23% accuracy on SEED, 73.82% on SEED-IV, and 76.38% on DEAP, achieving consistent improvements of 1.89%, 1.66%, and 1.28% over previous best methods, respectively.
