LEL: Lipschitz Continuity Constrained Ensemble Learning for Efficient EEG-Based Intra-subject Emotion Recognition

Accurate and efficient recognition of emotional states is critical for human social functioning, and impairments in this ability are associated with significant psychosocial difficulties. While electroencephalography (EEG) offers a powerful tool for objective emotion detection, existing EEG-based emotion recognition (EER) methods suffer from three key limitations: 1) insufficient model stability; 2) limited accuracy in processing high-dimensional nonlinear EEG signals; and 3) poor robustness against intrasubject variability and signal noise. To address these challenges, we introduce Lipschitz continuity-constrained ensemble learning (LEL), a novel framework that enhances EER by enforcing Lipschitz continuity constraints (LCCs) on transformer-based attention mechanisms, spectral extraction, and normalization modules. These heterogeneous constraints bound the global Lipschitz constant via function composition, ensures model stability, reduces sensitivity to signal variability and noise, and improves generalization capability. In addition, LEL uses a learnable ensemble fusion strategy that optimally combines decisions from multiple heterogeneous classifiers to mitigate single-model bias and variance. Extensive experiments on three public benchmark datasets (EAV, FACED, and SEED) demonstrate superior performance, achieving average recognition accuracies of 74.25%±2.3, 81.19%±2.8, and 86.79%±1.9, respectively. The official implementation codes are available at https://github.com/NZWANG/LEL