A Customized Convolutional Neural Network Design Using Improved Softmax Layer for Real-time Human Emotion Recognition
顯示簡易詮釋資料 統計資料| 欄位 | 內容 | 語言 |
|---|---|---|
dc.citation.epage |
106 | en_US |
dc.citation.spage |
102 | en_US |
dc.citation.woscount |
0 | en_US |
dc.contributor.author |
Wang, Kai-Yen | en_US |
dc.contributor.author |
Huang, Yu-De | en_US |
dc.contributor.author |
Ho, Yun-Lung | en_US |
dc.contributor.author |
Fang, Wai-Chi | en_US |
dc.contributor.department |
電子工程學系及電子研究所 | zh_TW |
dc.contributor.department |
Department of Electronics Engineering and Institute of Electronics | en_US |
dc.date.accessioned |
2019-12-13T01:12:51Z | |
dc.date.available |
2019-12-13T01:12:51Z | |
dc.date.issued |
2019-01-01 | en_US |
dc.description.abstract |
This paper proposes an improved softmax layer algorithm and hardware implementation, which is applicable to an effective convolutional neural network of EEG-based real-time human emotion recognition. Compared with the general softmax layer, this hardware design adds threshold layers to accelerate the training speed and replace the Euler's base value with a dynamic base value to improve the network accuracy. This work also shows a hardware-friendly way to implement batch normalization layer on chip. Using the EEG emotion DEAP[7] database, the maximum and mean classification accuracy were achieved as 96.03% and 83.88% respectively. In this work, the usage of improved softmax layer can save up to 15% of training model convergence time and also increase by 3 to 5% the average accuracy. | en_US |
dc.identifier.isbn |
978-1-5386-7884-8 | en_US |
dc.identifier.journal |
2019 IEEE INTERNATIONAL CONFERENCE ON ARTIFICIAL INTELLIGENCE CIRCUITS AND SYSTEMS (AICAS 2019) | en_US |
dc.identifier.uri |
https://ir.lib.nycu.edu.tw/handle/11536/153279 | |
dc.identifier.wosnumber |
WOS:000493095400023 | en_US |
dc.language.iso |
en_US | en_US |
dc.subject |
Improved Softmax Layer | en_US |
dc.subject |
Threshold Layer | en_US |
dc.subject |
Batch Normalization Layer | en_US |
dc.subject |
Convolutional Neural Network | en_US |
dc.subject |
Deep Learning | en_US |
dc.subject |
Hardware Machine Learning | en_US |
dc.title |
A Customized Convolutional Neural Network Design Using Improved Softmax Layer for Real-time Human Emotion Recognition | en_US |
dc.type |
Proceedings Paper | en_US |
檔案
本條目沒有附加檔案。
所屬典藏: 會議論文