Representation Enhancement for Convolutional Neural Network Using Filter Diversity

被引：0
|
作者：
Seo K. [1 ]
机构：
[1] Department of Electronics Engineering, Seokyeong University
来源：
Transactions of the Korean Institute of Electrical Engineers | 2022年 / 71卷 / 12期
关键词：
CNN; Deep learning; Feature Representation; Filter Diversity; Filter Spreading; Singular Value Decomposition (SVD) Entropy;
D O I：
10.5370/KIEE.2022.71.12.1825
中图分类号：
学科分类号：
摘要：
This paper aims to improve the feature representation by diversifying CNN filters inspired by niche concept in evolution. The singular value decomposition (SVD) entropy based efficient metric for diversity is proposed In the proposed approach, filters are clustered by groups and they are calculated as differences from the center values within the groups, rather than by entire rank based comparison. This provides an effective method for increasing the substantial diversity of filters. Furthermore, the filters with low diversity are adjusted by the diversity spreading framework for better diversity in the reconstruction process. The improvement of the filter representation by performing experiments on CIFAR 10/100 data for VGG16, and ImageNet for ResNet34 is provided. Because there are no similar studies, we compare our results with respect to those of relatively relevant pruning methods in terms of classification performance accuracy as well as the pruned rates and flops. Copyright © The Korean Institute of Electrical Engineers.
引用
收藏
页码：1825 / 1829
页数：4
相关论文
共 50 条
  • [11] The Enhancement of WiFi Fingerprint Positioning Using Convolutional Neural Network
    Zhang, Ting
    Man, Yi
    2018 INTERNATIONAL CONFERENCE ON COMPUTER, COMMUNICATION AND NETWORK TECHNOLOGY (CCNT 2018), 2018, 291 : 479 - 483
  • [12] Speech Enhancement using Convolutional Neural Network with Skip Connections
    Shi, Yupeng
    Rong, Weicong
    Zheng, Nengheng
    2018 11TH INTERNATIONAL SYMPOSIUM ON CHINESE SPOKEN LANGUAGE PROCESSING (ISCSLP), 2018, : 6 - 10
  • [13] Image Enhancement and Exposure Correction Using Convolutional Neural Network
    Parab M.
    Bhanushali A.
    Ingle P.
    Pavan Kumar B.N.
    SN Computer Science, 4 (2)
  • [14] Single channel speech enhancement using convolutional neural network
    Kounovsky, Tomas
    Malek, Jiri
    2017 IEEE INTERNATIONAL WORKSHOP OF ELECTRONICS, CONTROL, MEASUREMENT, SIGNALS AND THEIR APPLICATION TO MECHATRONICS (ECMSM), 2017,
  • [15] Locally Adaptive Contrast Enhancement Using Convolutional Neural Network
    Han, Bok Gyu
    Yang, Hyeon Seok
    Moon, Young Shik
    2018 IEEE INTERNATIONAL CONFERENCE ON CONSUMER ELECTRONICS (ICCE), 2018,
  • [16] Iterative PET Image Reconstruction Using Convolutional Neural Network Representation
    Gong, Kuang
    Guan, Jiahui
    Kim, Kyungsang
    Zhang, Xuezhu
    Yang, Jaewon
    Seo, Youngho
    El Fakhri, Georges
    Qi, Jinyi
    Li, Quanzheng
    IEEE TRANSACTIONS ON MEDICAL IMAGING, 2019, 38 (03) : 675 - 685
  • [17] LEARNING A REPRESENTATION FOR COVER SONG IDENTIFICATION USING CONVOLUTIONAL NEURAL NETWORK
    Yu, Zhesong
    Xu, Xiaoshuo
    Chen, Xiaoou
    Yang, Deshun
    2020 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH, AND SIGNAL PROCESSING, 2020, : 541 - 545
  • [18] Malware Traffic Classification Using Convolutional Neural Network for Representation Learning
    Wang, Wei
    Zhu, Ming
    Zeng, Xuewen
    Ye, Xiaozhou
    Sheng, Yiqiang
    2017 31ST INTERNATIONAL CONFERENCE ON INFORMATION NETWORKING (ICOIN), 2017, : 712 - 717
  • [19] A chemically inspired convolutional neural network using electronic structure representation
    Mok, Dong Hyeon
    Shin, Daeun
    Na, Jonggeol
    Back, Seoin
    JOURNAL OF MATERIALS CHEMISTRY A, 2023, 11 (19)
  • [20] Ear Detection Using Convolutional Neural Network on Graphs with Filter Rotation
    Tomczyk, Arkadiusz
    Szczepaniak, Piotr S.
    SENSORS, 2019, 19 (24)
12345