跳至主導覽
跳至搜尋
跳過主要內容
國立中央大學 首頁
說明與常見問題
English
中文
首頁
人才檔案
研究單位
研究成果
資料集
榮譽/獲獎
學術活動
新聞/媒體
影響
按專業知識、姓名或所屬機構搜尋
ReActXGB: A Hybrid Binary Convolutional Neural Network Architecture for Improved Performance and Computational Efficiency
Po Hsun Chu,
Ching Han Chen
資訊工程學系
研究成果
:
書貢獻/報告類型
›
會議論文篇章
›
同行評審
總覽
指紋
指紋
深入研究「ReActXGB: A Hybrid Binary Convolutional Neural Network Architecture for Improved Performance and Computational Efficiency」主題。共同形成了獨特的指紋。
排序方式
重量
按字母排序
Keyphrases
Computational Efficiency
100%
Performance Improvement
100%
Performance Efficiency
100%
Convolutional Neural Network Architecture
100%
Binary Convolutional Neural Network
100%
Fully Connected Layer
50%
Low Computational Complexity
25%
Hybrid Model
25%
Computational Cost
25%
Memory Requirements
25%
Computational Resources
25%
Deep Neural Network
25%
Binarization
25%
XGBoost
25%
Floating-point Operations
25%
Performance Gap
25%
Model Size
25%
Input Layer
25%
Value Network
25%
Conventional Practice
25%
Fashion-MNIST
25%
Fully Convolutional Layer
25%
Computer Science
Computational Efficiency
100%
Neural Network Architecture
100%
Convolutional Neural Network
100%
Computational Cost
50%
Fully Connected Layer
50%
Experimental Result
25%
Memory Requirement
25%
Computational Resource
25%
Deep Neural Network
25%
Extreme Gradient Boosting
25%
Convolutional Layer
25%
Potential Solution
25%
Performance Gap
25%
binarization
25%
Computational Bottleneck
25%
Floating-Point Operation
25%
Engineering
Computational Efficiency
100%
Neural Network Architecture
100%
Convolutional Neural Network
100%
Computational Cost
50%
Experimental Result
25%
Hybrid Model
25%
Memory Requirement
25%
Floating Point
25%
Point Operation
25%
Convolutional Layer
25%
Computational Resource
25%
Input Layer
25%
Deep Neural Network
25%
Chemical Engineering
Neural Network
100%
Deep Neural Network
25%
Biochemistry, Genetics and Molecular Biology
Solution and Solubility
100%