首页|基于域间Mixup微调策略的跨被试运动想象脑电信号分类算法

基于域间Mixup微调策略的跨被试运动想象脑电信号分类算法

扫码查看
原文链接
  • 万方数据
  • 维普
为了缓解传统微调算法的灾难性遗忘问题,本文提出了一种基于域间Mixup微调策略的跨被试运动想象脑电信号分类算法Mix-Tuning.Mix-Tuning通过预训练、微调的二阶段训练方式,实现跨领域知识迁移.预训练阶段,Mix-Tuning使用源域数据初始化模型参数,挖掘源域数据潜在信息.微调阶段,Mix-Tuning通过域间Mixup,生成域间插值数据微调模型参数.域间Mixup数据增强策略引入源域数据潜在信息,缓解传统微调算法在样本稀疏场景下的灾难性遗忘问题,提高模型的泛化性能.Mix-Tuning被进一步应用于运动想象脑电信号分类任务,实现了跨被试正向知识迁移.Mix-Tuning在BMI数据集的运动想象任务达到了 85.50%的平均分类准确率,相较于被试-依赖和被试-独立训练方式的预测准确率 58.72%和 84.01%,分别提高26.78%和1.49%.本文分析结果可为跨被试运动想象脑电信号分类算法提供参考.
Cross-subject motor imagery EEG classification based on inter-domain Mixup fine-tuning strategy
In order to alleviate the catastrophic forgetting problem of vanilla fine-tuning algorithms,we propose a cross-subject motor imagery EEG classification method based on inter-domain Mixup fine-tuning strategy,i.e.,Mix-Tuning.Mix-Tuning realizes cross-domain knowledge transfer through a two-stage training manner consisting of pre-training and fine-tuning.In the pre-training stage,Mix-Tuning uses the source domain data to initialize the model parameters and mine potential information of the source domain data.In the fine-tuning stage,Mix-Tuning generates inter-domain inter-polation data to fine-tune the model parameters through inter-domain Mixup.Inter-domain Mixup data enhancement strategy introduces latent information of the source domain data,which alleviates the catastrophic forgetting problem of Vanilla Fine-tuning in sparse sample scenarios and improves the generalization performance of the model.Mix-Tuning is further applied to the motor imagery EEG classification task and achieves cross-subject positive knowledge transfer.Mix-Tuning achieved an average classification accuracy of 85.50%on motor imagery task BMIdataset.Compared with 58.72%and 84.01%for Subject-specific and Subject-independent training manner,Mix-Tuning increased by 26.78%and 1.49%,respectively.The analysis results in this paper can provide a reference for cross-subject motor imagery EEG classification algorithm.

inter-domain Mixuppre-trainingfine-tuningelectroencephalogrammotor imagerycross-subject know-ledge transferconvolutional neural networkregularization

蒋云良、周阳、张雄涛、苗敏敏、张永

展开 >

湖州师范学院 信息工程学院,浙江 湖州 313000

湖州师范学院 浙江省现代农业资源智慧管理与应用研究重点实验室,浙江 湖州 313000

浙江师范大学 计算机科学与技术学院,浙江 金华 321000

域间Mixup 预训练 微调 脑电信号 运动想象 跨被试知识迁移 卷积神经网络 正则化

国家自然科学基金项目国家自然科学基金项目国家自然科学基金项目国家自然科学基金项目浙江省教育厅科研项目

617711936210118962376094U22A20102Y202146028

2024

10.11992/tis.202208017

智能系统学报
中国人工智能学会　哈尔滨工程大学

智能系统学报

CSTPCD北大核心
影响因子：0.672
ISSN：1673-4785
年,卷(期)：2024.19(4)
  • 2