计算机科学

计算机科学 ›› 2026, Vol. 53 ›› Issue (8): 94-102.doi: 10.11896/jsjkx.250700141

• 数据库 & 大数据 & 数据科学 • 上一篇    下一篇

基于代理的无源域适应脑电情绪识别方法

吴水清, 邱纪豪, 刘祥, 董昱希, 文益民   

  1. 桂林电子科技大学广西图像图形与智能处理重点实验室 广西 桂林 541004
  • 收稿日期:2025-07-21 修回日期:2025-11-24 出版日期:2026-08-15 发布日期:2026-08-17
  • 通讯作者: 刘祥(21032303090@mails.guet.edu.cn)
  • 作者简介:(wsqsix@163.com)
  • 基金资助:
    广西自然科学基金重点项目(2024GXNSFDA010066);国家自然科学基金(62366011);广西学位与研究生教育改革课题(JGY2023128);桂林电子科技大学研究生教育创新计划(2024YCXS038)

Proxy-based Source-free Domain Adaptation for EEG Emotion Recognition Method

WU Shuiqing, QIU Jihao, LIU Xiang, DONG Yuxi, WEN Yimin   

  1. Guangxi Key Laboratory of Image and Graphic Intelligent Processing, Guilin University of Electronic Technology, Guilin, Guangxi 541004, China
  • Received:2025-07-21 Revised:2025-11-24 Published:2026-08-15 Online:2026-08-17
  • About author:WU Shuiqing,born in 2001,postgra-duate.Her main research interests include transfer learning and EEG emotion recognition.
    LIU Xiang,born in 1995,Ph.D.His main research interests include machine learning and streaming data mining.
  • Supported by:
    Key Project of Guangxi Natural Science Foundation(2024GXNSFDA010066), National Natural Science Foundation of China(62366011),Innovation Project of Guangxi Graduate Education(JGY2023128) and Innovation Project of GUET Graduate Education(2024YCXS038).

PDF (PC)

112

摘要： 在情感脑机接口领域,基于脑电信号的情绪识别取得了显著进展。然而,传统无监督域适应方法通常需要同时访问源域与目标域数据,存在源域受试者隐私泄露的风险。为了在无需源域样本的情况下估计域间差异并抑制噪声伪标签,提出了一种基于代理的无源域适应脑电情绪识别方法。该方法以预训练源模型分类器权重矩阵的行向量作为类原型,据此筛选目标域中的最近邻样本,构建类平衡的代理源域;进而利用该代理源域训练目标模型,并通过优化类原型与样本筛选过程提升代理域质量。此外,采用Mixup算法混合特征提取后的目标域特征,以提升特征表示能力;并提出伪标签加权校正策略,通过不确定性估计对分类损失重新加权与校正不确定性高的样本的伪标签。与主流的无源域适应方法相比,所提方法在SEED,SEED-IV与SEED-V数据集上的准确率平均提升了3.24%,同时在伪标签质量方面表现出显著优势1)。

关键词: 脑电信号, 情绪识别, 无源域适应, 隐私保护, 深度学习

Abstract: Emotion recognition using electroencephalography(EEG) has made significant progress in the field of emotional brain-computer interfaces.However,traditional unsupervised domain adaptation(UDA) methods typically require simultaneous access to both source and target domain data,posing a risk of privacy leakage for source domain subjects.To estimate inter-domain differences and suppress noisy pseudo-labels without the need for source domain samples,this paper proposes a proxy-based source-free domain adaptation for EEG emotion recognition method.This method uses the row vectors of the classifier weight matrix from a pre-trained source model as class prototypes,based on which the nearest neighbor samples in the target domain are selected to construct a class-balanced proxy source domain.Subsequently,this proxy source domain is used to train the target model,and the quality of the proxy domain is improved by optimizing the class prototypes and sample selection process.Addi-tionally,the Mixup algorithm is employed to mix the features extracted from the target domain to enhance feature representation capability.A pseudo-label weighting and correction strategy is also proposed,which re-weights the classification loss through uncertainty estimation and corrects the pseudo-labels of samples with high uncertainty.Compared to mainstream source-free domain adaptation(SFDA) methods,the proposed approach achieves an average accuracy improvement of 3.24% on the SEED,SEED-IV,and SEED-V datasets,while also demonstrating significant advantages in pseudo-label quality.

Key words: Electroencephalogram, Emotion recognition, Source-free domain adaptation, Privacy-preserving, Deep learning

中图分类号: 

  • R318
[1] YIN H L,ZHENG W L,LU B L.STAR:A Spatial-TemporalAutoencoder for EEG Restoration in Emotion Recognition[C]//Proceedings of the ICASSP 2025-2025 IEEE International Conference on Acoustics,Speech and Signal Processing(ICASSP).IEEE,2025:1-5.
[2] BALLESTEROS J A,RAMÍREZ V G M,MOREIRA F,et al.Facial emotion recognition through artificial intelligence[J].Frontiers in Computer Science,2024,6:1359471.
[3] GARBER-BARRON M,SI M.Using body movement and posture for emotion detection in non-acted scenarios[C]//Procee-dings of the 2012 IEEE International Conference on Fuzzy Systems.IEEE,2012:1-8.
[4] BÄNZIGER T,GRANDJEAN D,SCHERER K R.Emotion re-cognition from expressions in face,voice,and body:the Multimodal Emotion Recognition Test(MERT)[J].Emotion,2009,9(5):691-704.
[5] SAMARA A,MENEZES M L R,GALWAY L.Feature extraction for emotion recognition and modelling using neurophysiological data[C]//Proceedings of the 2016 15th International Conference on Ubiquitous Computing and Communications and 2016 International Symposium on Cyberspace and Security(IUCC-CSS).IEEE,2016:138-144.
[6] LAN Y T,JIANG W B,ZHENG W L,et al.CEMOAE:A dynamic autoencoder with masked channel modeling for robustEEG-based emotion recognition[C]//Proceedings of the ICASSP 2024-2024 IEEE International Conference on Acoustics,Speech and Signal Processing(ICASSP).IEEE,2024:1871-1875.
[7] WANG Y,LIU J W,LU B L,et al.From EEG toeye movements:cross-modal emotion recognition using constrained adversarial network with dual attention[J].IEEE Transactions on Affective Computing,2025,16(3):1543-1556.
[8] AN Y,HU S,LIU S,et al.Cross-subject EEG emotion recognition based on interconnected dynamic domain adaptation[C]//Proceedings of the ICASSP 2024-2024 IEEE International Conference on Acoustics,Speech and Signal Processing(ICASSP).IEEE,2024:12981-12985.
[9] LI X,CHEN C L P,CHEN B,et al.Gusa:graph-based unsupervised subdomain adaptation for cross-subject EEG emotion recognition[J].IEEE Transactions on Affective Computing,2024,15(3):1451-1462.
[10] ZHANG Y,CHEN S,JIANG W,et al.Domain-guided condi-tional diffusion model for unsupervised domain adaptation[J].Neural Networks,2025,184:107031.
[11] LI R,JIAO Q,CAO W,et al.Model adaptation:unsupervised domain adaptation without source data[C]//Proceedings of the 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition(CVPR).IEEE,2020:9638-9647.
[12] LIANG J,HU D,FENG J.Do we really need to access thesource data? source hypothesis transfer for unsupervised domain adaptation[C]//Proceedings of the International Conference on Machine Learning.PMLR,2020:6028-6039.
[13] LIANG J,HU D,WANG Y,et al.Source data-absent unsupervised domain adaptation through hypothesis transfer and labeling transfer[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2021,44(11):8602-8617.
[14] SAITO K,WATANABE K,USHIKU Y,et al.Maximum classifier discrepancy for unsupervised domain adaptation[C]//Proceedings of the 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition(CVPR).IEEE,2018:3723-3732.
[15] ZHANG Q,ZHANG J,LIU W,et al.Category anchor-guided unsupervised domain adaptation for semantic segmentation[C]//Proceedings of the 33rd International Conference on Neural Information Processing Systems.Curran Associates Inc.,2019:435-445.
[16] CHAI X,WANG Q,ZHAO Y,et al.Unsupervised domain adaptation techniques based on auto-encoder for non-stationary EEG-based emotion recognition[J].Computers in Biology and Medicine,2016,79:205-214.
[17] JIMÉNEZ-GUARNEROS M,FUENTES-PINEDA G.Lear-ning a robust unified domain adaptation framework for cross-subject EEG-based emotion recognition[J].Biomedical Signal Processing and Control,2023,86:105138.
[18] LI J,YU Z,DU Z,et al.A comprehensive survey on source-free domain adaptation[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2024,46(8):5743-5762.
[19] HUANG J,GUAN D,XIAO A,et al.Model adaptation:historical contrastive learning for unsupervised domain adaptation without source data[C]//Proceedings of the Advances in Neural Information Processing Systems.Curran Associates Inc.,2021,34:3635-3649.
[20] TIAN J,ZHANG J,LI W,et al.VDM-DA:virtual domain mo-deling for source data-free domain adaptation[J].IEEE Transactions on Circuits and Systems for Video Technology,2022,32(6):3749-3760.
[21] ZHAO H Y,LI C,LIU Y.EEG Emotion Recognition Based on Source-Free Domain Adaptation[J].Chinese Journal of Biome-dical Engineering,2024,43(2):129-142.
[22] SALIMNIA A H.Attention-based Multi-Source-Free Domain Adaptation for EEG Emotion Recognition[D].Western:The University of Western Ontario,2023.
[23] LIU Z,CHEN G,LI Z,et al.PSDC:a prototype-based shared-dummy classifier model for open-set domain adaptation[J].IEEE Transactions on Cybernetics,2023,53(11):7353-7366.
[24] CHEN W Y,LIU Y C,KIRA Z,et al.A closer look at few-shot classification[J].arXiv:1904.04232,2019.
[25] DU Y,YANG H,CHEN M,et al.Generation,augmentation,and alignment:a pseudo-source domain based method for source-free domain adaptation[J].Machine Learning,2024,113(6):3611-3631.
[26] ZHANG H,CISSE M,DAUPHIN Y N,et al.mixup:Beyond empirical risk minimization[J].arXiv:1710.09412,2017.
[27] XIE J,GIRSHICK R,FARHADI A.Unsupervised deep embedding for clustering analysis[C]//Proceedings of the 33rd International Conference on Machine Learning.PMLR,2016:478-487.
[28] LIANG J,HU D,FENG J.Domain adaptation with auxiliarytarget domain-oriented classifier[C]//Proceedings of the 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition(CVPR).IEEE,2021:16627-16637.
[29] DING Y,SHENG L,LIANG J,et al.ProxyMix:proxy-based mixup training with label refinery for source-free domain adaptation[J].Neural Networks,2023,167:92-103.
[30] DUAN R N,ZHU J Y,LU B L.Differential entropy feature for EEG-based emotion classification[C]//Proceedings of the 2013 6th International IEEE/EMBS Conference on Neural Enginee-ring(NER).IEEE,2013:81-84.
[31] ZHENG W L,LIU W,LU Y,et al.EmotionMeter:A multimodal framework for recognizing human emotions[J].IEEE Transactions on Cybernetics,2019,49(3):1110-1122.
[32] LIU W,QIU J L,ZHENG W L,et al.Comparing recognitionperformance and robustness of multimodal deep learning models for multimodal emotion recognition[J].IEEE Transactions on Cognitive and Developmental Systems,2022,14(2):715-729.
[33] LEE J,JUNG D,YIM J,et al.Confidence score for source-free unsupervised domain adaptation[C]//Proceedings of the International Conference on Machine Learning.PMLR,2022:12365-12377.
[34] KUMAR V,LAL R,PATIL H,et al.Conmix for source-free single and multi-target domain adaptation[C]//Proceedings of the 2023 IEEE/CVF Winter Conference on Applications of Computer Vision(WACV).IEEE,2023:4167-4177.
[35] HE J,WU L,TAO C,et al.Source-free domain adaptation with unrestricted source hypothesis[J].Pattern Recognition,2024,149:110246.
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
No Suggested Reading articles found!
渝ICP备16013121号-4
地址：重庆市两江新区洪湖西路18号    邮编：401121  
电话：023-63500828（编辑部） 023-67039612（会议/专题） 023-67039623（财务/发票）
E-mail：jsjkx12@163.com
本系统由北京玛格泰克科技发展有限公司设计开发