FusionGen: Feature Fusion-Based Few-Shot EEG Data Generation

2025-10-12

AI Summary

To address the challenges of limited electroencephalography (EEG) data and substantial inter-subject variability in brain–computer interfaces (BCIs), which severely hinder the generalizability of decoding models, this paper proposes FusionGen—a novel framework for high-fidelity EEG generation under few-shot conditions. FusionGen decouples representation learning from feature fusion and introduces a feature-matching fusion module that jointly performs cross-trial feature integration and lightweight reconstruction, thereby enhancing both generative diversity and model trainability. Extensive experiments on multiple public EEG datasets demonstrate that FusionGen significantly outperforms existing data augmentation methods, achieving average classification accuracy improvements of 3.2–5.8 percentage points. By effectively mitigating subject-specific variations, FusionGen strengthens the robustness and generalization capability of downstream decoding models.

Technology Category

• Machine Learning: Deep Generative Models & Autoencoders
• Natural Language Processing: Generation
• Cognitive Modeling & Cognitive Systems: Neural Spike Coding

Application Category

• Economics, Online Markets and Human Computation: Economic ramifications for generative AI infrastructure and applications
• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Brain-computer interfaces (BCIs) provide potential for applications ranging from medical rehabilitation to cognitive state assessment by establishing direct communication pathways between the brain and external devices via electroencephalography (EEG). However, EEG-based BCIs are severely constrained by data scarcity and significant inter-subject variability, which hinder the generalization and applicability of EEG decoding models in practical settings. To address these challenges, we propose FusionGen, a novel EEG data generation framework based on disentangled representation learning and feature fusion. By integrating features across trials through a feature matching fusion module and combining them with a lightweight feature extraction and reconstruction pipeline, FusionGen ensures both data diversity and trainability under limited data constraints. Extensive experiments on multiple publicly available EEG datasets demonstrate that FusionGen significantly outperforms existing augmentation techniques, yielding notable improvements in classification accuracy.

Problem

• Addresses EEG data scarcity in brain-computer interfaces
• Reduces inter-subject variability in EEG decoding models
• Enhances generalization of EEG models with limited data

Innovation

• Feature fusion-based EEG data generation framework
• Disentangled representation learning for data diversity
• Lightweight feature extraction and reconstruction pipeline