• Home
  • Search
  • Adaptive deep feature representation learning for cross-subject EEG decoding
  • Cite Icon11
  • https://doi.org/10.1186/s12859-024-06024-wCopy DOI Icon

Adaptive deep feature representation learning for cross-subject EEG decoding

Show More
  • Abstract
  • Literature Map
  • References
  • Citations
  • Similar Papers
Abstract

Background:The collection of substantial amounts of electroencephalogram (EEG) data is typically time-consuming and labor-intensive, which adversely impacts the development of decoding models with strong generalizability, particularly when the available data is limited. Utilizing sufficient EEG data from other subjects to aid in modeling the target subject presents a potential solution, commonly referred to as domain adaptation. Most current domain adaptation techniques for EEG decoding primarily focus on learning shared feature representations through domain alignment strategies. Since the domain shift cannot be completely removed, target EEG samples located near the edge of clusters are also susceptible to misclassification.Methods:We propose a novel adaptive deep feature representation (ADFR) framework to improve the cross-subject EEG classification performance through learning transferable EEG feature representations. Specifically, we first minimize the distribution discrepancy between the source and target domains by employing maximum mean discrepancy (MMD) regularization, which aids in learning the shared feature representations. We then utilize the instance-based discriminative feature learning (IDFL) regularization to make the learned feature representations more discriminative. Finally, the entropy minimization (EM) regularization is further integrated to adjust the classifier to pass through the low-density region between clusters. The synergistic learning between above regularizations during the training process enhances EEG decoding performance across subjects.Results:The effectiveness of the ADFR framework was evaluated on two public motor imagery (MI)-based EEG datasets: BCI Competition III dataset 4a and BCI Competition IV dataset 2a. In terms of average accuracy, ADFR achieved improvements of 3.0% and 2.1%, respectively, over the state-of-the-art methods on these datasets.Conclusions:The promising results highlight the effectiveness of the ADFR algorithm for EEG decoding and show its potential for practical applications.

Similar Papers
  • Research Article
  • Citations42

Enhanced hierarchical symbolic dynamic entropy and maximum mean and covariance discrepancy-based transfer joint matching with Welsh loss for intelligent cross-domain bearing health monitoring

  • Sep 02, 2021
  • Mechanical Systems and Signal Processing
  • Cheng Yang +3
  • Research Article
  • Citations48

Maximum Mean and Covariance Discrepancy for Unsupervised Domain Adaptation

  • Aug 08, 2019
  • Neural Processing Letters
  • Wenju Zhang +3
  • Research Article
  • Citations5

Learning transferable and discriminative features for unsupervised domain adaptation

  • Mar 14, 2022
  • Intelligent Data Analysis
  • Yuntao Du +5
  • Book Chapter
  • Citations2

Unsupervised Domain Adaptation for Object Detection Using Distribution Matching in Various Feature Level

  • Jan 01, 2019
  • Hyoungwoo Park +3
  • Research Article
  • Citations20

Priming cross-session motor imagery classification with a universal deep domain adaptation framework

  • Aug 16, 2023
  • Neurocomputing
  • Xin Zhang +5
  • Research Article
  • Citations15

Time–space–frequency feature Fusion for 3-channel motor imagery classification

  • Dec 17, 2023
  • Biomedical Signal Processing and Control
  • Zhengqing Miao +1
  • Research Article
  • Citations5

Multi-source deep domain adaptation ensemble framework for cross-dataset motor imagery EEG transfer learning

  • May 01, 2024
  • Physiological Measurement
  • Minmin Miao +5
  • Research Article
  • Citations13

Explainable cross-task adaptive transfer learning for motor imagery EEG classification

  • Nov 30, 2023
  • Journal of Neural Engineering
  • Minmin Miao +5
  • Research Article
  • Citations1

Deep Learning-Based Financial Fraud Detection with Temporal and Feature-Level Adaptation

  • Aug 29, 2025
  • Journal of Computing and Electronic Information Management
  • Weilun Tsai
  • Dissertation
  • Citations1

SEISMIC FAULT SEGMENTATION USING UNSUPERVISED DOMAIN ADAPTATION

  • Aug 10, 2023
  • Maykol Jiampiers Campos Trinidad
  • Research Article
  • Citations2

Discriminative possibilistic clustering promoting cross-domain emotion recognition.

  • Nov 01, 2024
  • Frontiers in neuroscience
  • Yufang Dan +2
  • Research Article
  • Citations91

Mutual Information-Driven Subject-Invariant and Class-Relevant Deep Representation Learning in BCI.

  • Feb 01, 2023
  • IEEE Transactions on Neural Networks and Learning Systems
  • Eunjin Jeon +3
  • Research Article
  • Citations26

Selective multi–view time–frequency decomposed spatial feature matrix for motor imagery EEG classification

  • Jan 24, 2024
  • Expert Systems with Applications
  • Tian-Jian Luo
  • Research Article
  • Citations6

Unsupervised Domain Adaptation With Synchronized Self-Training for Cross- Domain Motor Imagery Recognition.

  • May 01, 2025
  • IEEE journal of biomedical and health informatics
  • Peiyin Chen +7
  • Book Chapter
  • Citations3

Agnostic Domain Adaptation

  • Jan 01, 2011
  • Alexander Vezhnevets +1
Cactus Communications logo

Copyright 2026 Cactus Communications. All rights reserved.