EEG-based emotion recognition is critical for mental health monitoring and affective brain-computer interfaces, yet existing deep learning approaches often treat emotion classes as isolated labels, ignoring their psychological interdependencies. We propose a graph-regularized learning framework that conceptualizes emotions as nodes in a graph where edges encode proximity based on dimensional emotion theories.
We adopt three complementary regularization strategies:
- Graph Label Smoothing (intuitive soft labeling)
- Commuting distance on graph via Graph Laplacian (spectral graph theory)
- Sliced Wasserstein Distance (optimal transport on graph)
These strategies are ordered by increasing computational complexity and penalize model predictions that deviate from the established emotion topology. Our framework is evaluated across three representative backbone architectures: AudioTransformer (pure transformer), Conformer (CNN-transformer hybrid), and DCGNN (causal graph neural network), demonstrating architecture-agnostic benefits.
Experiments on SEED-IV (4 classes) and SEED-V (5 classes) datasets show consistent improvements: best case up to +5.42% accuracy and 39% reduction in psychologically implausible misclassifications. Ultimately, our framework helps raise the upper bound of performance achievable with standard approaches. Code will be released.
Blogger's Review: This research successfully enhances EEG emotion recognition accuracy by introducing graph regularization methods, showcasing the deep connections between emotions. Future studies could explore the application potential of this approach in other domains.