One Model for All: Universal Pre-training for EEG based Emotion Recognition across Heterogeneous Datasets and Paradigms

TL;DR

This paper introduces a universal pre-training framework for EEG-based emotion recognition that effectively handles dataset heterogeneity, achieving state-of-the-art results across multiple benchmarks and enabling robust cross-dataset transfer.

Contribution

The authors propose a novel two-stage pre-training approach with a unified channel schema and a new architecture combining ART and GAT, significantly improving generalization and transferability.

Findings

Universal pre-training stabilizes training and improves performance.

Achieves new state-of-the-art on SEED, DEAP, and DREAMER datasets.

Demonstrates effective cross-dataset transfer with high accuracy.

Abstract

EEG-based emotion recognition is hampered by profound dataset heterogeneity (channel/subject variability), hindering generalizable models. Existing approaches struggle to transfer knowledge effectively. We propose 'One Model for All', a universal pre-training framework for EEG analysis across disparate datasets. Our paradigm decouples learning into two stages: (1) Univariate pre-training via self-supervised contrastive learning on individual channels, enabled by a Unified Channel Schema (UCS) that leverages the channel union (e.g., SEED-62ch, DEAP-32ch); (2) Multivariate fine-tuning with a novel 'ART' (Adaptive Resampling Transformer) and 'GAT' (Graph Attention Network) architecture to capture complex spatio-temporal dependencies. Experiments show universal pre-training is an essential stabilizer, preventing collapse on SEED (vs. scratch) and yielding substantial gains on DEAP (+7.65%) and DREAMER (+3.55%). Our framework achieves new SOTA performance on all within-subject benchmarks: SEED (99.27%), DEAP (93.69%), and DREAMER (93.93%). We also show SOTA cross-dataset transfer, achieving 94.08% (intersection) and 93.05% (UCS) on the unseen DREAMER dataset, with the former surpassing the within-domain pre-training benchmark. Ablation studies validate our architecture: the GAT module is critical, yielding a +22.19% gain over GCN on the high-noise DEAP dataset, and its removal causes a catastrophic -16.44% performance drop. This work paves the way for more universal, scalable, and effective pre-trained models for diverse EEG analysis tasks.