大工至善|大学至真分享 http://blog.sciencenet.cn/u/lcj2212916

博文

[转载]【计算机科学】【2017.08】脑电信号分类中的深度学习与迁移学习

已有 321 次阅读 2019-5-19 16:16 |系统分类:科研笔记|文章来源:转载


本文为美国内布拉斯加大学作者:Jacob M. Williams)的硕士论文82

 

尽管在其他多种空间和时间序列数据中实现了最先进的分类精度但深度学习很少用于脑电图(EEG)信号的分类相反大多数研究继续使用手工特征提取然后使用传统的分类器SVMs或逻辑回归这主要是由于每个实验的样本数量少数据的高维性质以及难以找到适合于脑电图信号分类的深度学习架构本文将几种深度学习架构与传统的视觉诱发脑电信号分类方法进行了比较我们发现使用长短期记忆单元(LSTM)的深度学习体系架构优于传统方法而小型卷积体系架构的性能与传统方法相当我们还探讨了通过跨多个科目的训练和对特定科目的改进来使用迁移学习这种形式的迁移学习进一步提高了深度学习模型的分类精度

 

Deep learning is seldom used in the classification of electroencephalography (EEG) signals, despite achieving state of the art classification accuracies in other spatial and time series data. Instead, most research has continued to use manual feature extraction followed by a traditional classifier, such as SVMs or logistic regression. This is largely due to the low number of samples per experiment, high-dimensional nature of the data, and the difficulty in finding appropriate deep learning architectures for classification of EEG signals. In this thesis, several deep learning architectures are compared to traditional techniques for the classification of visually evoked EEG signals. We found that deep learning architectures using long short-term memory units (LSTMs) outperform traditional methods, while small convolutional architectures performed comparably to traditional methods. We also explored the use of transfer learning by training across multiple subjects and refining on a particular subject. This form of transfer learning further improved the classification accuracy of the deep learning models.

 

引言

相关工作与研究背景

问题定义与使用方法

数据集具体实现与结果

总结与未来工作展望 


下载英文原文地址:

http://page2.dfpan.com/fs/7lcje221f291d650b72/ 


更多精彩文章请关注微信号:qrcode_for_gh_60b944f6c215_258.jpg



http://blog.sciencenet.cn/blog-69686-1179973.html

上一篇:[转载]【新书推荐】【2018】用于低强度冲突和国土安全的光电设备
下一篇:[转载]【源码】基于SIMULINK的LEGO Mindstorms EV3 BallBot机器人

0

该博文允许注册用户评论 请点击登录 评论 (0 个评论)

数据加载中...
扫一扫,分享此博文

Archiver|手机版|科学网 ( 京ICP备14006957 )

GMT+8, 2019-6-26 16:27

Powered by ScienceNet.cn

Copyright © 2007- 中国科学报社

返回顶部