Adaptive Split-MMD Training for Small-Sample Cross-Dataset P300 EEG Classification

2025-10-24

AI Summary

To address performance degradation in few-shot cross-dataset P300 EEG classification caused by source–target domain distribution shift, this paper proposes a backbone-agnostic adaptive domain adaptation framework. Methodologically, it introduces: (1) adaptive segmented Maximum Mean Discrepancy (MMD) alignment using a parameter-free logarithmic-layer MMD and median-heuristic RBF kernel; (2) target-weighted loss coupled with split-domain batch normalization (Split-BN) sharing affine parameters; and (3) end-to-end training via EEG Conformer architecture with square-root-scaled learning rate warmup. Evaluated on Active Visual Oddball and ERP CORE P3 datasets, the method achieves accuracies of 0.66 and 0.61, and AUCs of 0.74 and 0.65, respectively—significantly outperforming both target-only and mixed-training baselines. Ablation studies confirm the efficacy of each component.

Technology Category

• Machine Learning: Transfer, Domain Adaptation, Multi-Task Learning
• Cognitive Modeling & Cognitive Systems: Adaptive Behavior
• Computer Vision: Adversarial Attacks & Robustness

Application Category

• Search and Retrieval-Augmented AI: Vertical and domain-specific search
• User Modeling, Personalization and Recommendation: User modeling for targeted and personalized online advertising
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs

Abstract

Detecting single-trial P300 from EEG is difficult when only a few labeled trials are available. When attempting to boost a small target set with a large source dataset through transfer learning, cross-dataset shift arises. To address this challenge, we study transfer between two public visual-oddball ERP datasets using five shared electrodes (Fz, Pz, P3, P4, Oz) under a strict small-sample regime (target: 10 trials/subject; source: 80 trials/subject). We introduce Adaptive Split Maximum Mean Discrepancy Training (AS-MMD), which combines (i) a target-weighted loss with warm-up tied to the square root of the source/target size ratio, (ii) Split Batch Normalization (Split-BN) with shared affine parameters and per-domain running statistics, and (iii) a parameter-free logit-level Radial Basis Function kernel Maximum Mean Discrepancy (RBF-MMD) term using the median-bandwidth heuristic. Implemented on an EEG Conformer, AS-MMD is backbone-agnostic and leaves the inference-time model unchanged. Across both transfer directions, it outperforms target-only and pooled training (Active Visual Oddball: accuracy/AUC 0.66/0.74; ERP CORE P3: 0.61/0.65), with gains over pooling significant under corrected paired t-tests. Ablations attribute improvements to all three components.

Problem

• Addressing cross-dataset shift in small-sample P300 EEG classification
• Improving transfer learning between visual-oddball ERP datasets
• Enhancing single-trial P300 detection with limited labeled trials

Innovation

• Adaptive Split-MMD training for cross-dataset EEG classification
• Target-weighted loss with warm-up and Split Batch Normalization
• Parameter-free RBF-MMD term with median-bandwidth heuristic