TY - GEN
T1 - Exploiting convolution filter patterns for transfer learning
AU - Aygün, Mehmet
AU - Aytar, Yusuf
AU - Ekenel, Hazim Kemal
N1 - Publisher Copyright:
© 2017 IEEE.
PY - 2017/7/1
Y1 - 2017/7/1
N2 - In this paper, we introduce a new regularization technique for transfer learning. The aim of the proposed approach is to capture statistical relationships among convolution filters learned from a well-Trained network and transfer this knowledge to another network. Since convolution filters of the prevalent deep Convolutional Neural Network (CNN) models share a number of similar patterns, in order to speed up the learning procedure, we capture such correlations by Gaussian Mixture Models (GMMs) and transfer them using a regularization term. We have conducted extensive experiments on the CIFAR10, Places2, and CM-Places datasets to assess generalizability, task transferability, and cross-model transferability of the proposed approach, respectively. The experimental results show that the feature representations have efficiently been learned and transferred through the proposed statistical regularization scheme. Moreover, our method is an architecture independent approach, which is applicable for a variety of CNN architectures.
AB - In this paper, we introduce a new regularization technique for transfer learning. The aim of the proposed approach is to capture statistical relationships among convolution filters learned from a well-Trained network and transfer this knowledge to another network. Since convolution filters of the prevalent deep Convolutional Neural Network (CNN) models share a number of similar patterns, in order to speed up the learning procedure, we capture such correlations by Gaussian Mixture Models (GMMs) and transfer them using a regularization term. We have conducted extensive experiments on the CIFAR10, Places2, and CM-Places datasets to assess generalizability, task transferability, and cross-model transferability of the proposed approach, respectively. The experimental results show that the feature representations have efficiently been learned and transferred through the proposed statistical regularization scheme. Moreover, our method is an architecture independent approach, which is applicable for a variety of CNN architectures.
UR - http://www.scopus.com/inward/record.url?scp=85046246124&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85046246124&partnerID=8YFLogxK
U2 - 10.1109/ICCVW.2017.309
DO - 10.1109/ICCVW.2017.309
M3 - Conference contribution
AN - SCOPUS:85046246124
T3 - Proceedings - 2017 IEEE International Conference on Computer Vision Workshops, ICCVW 2017
SP - 2674
EP - 2680
BT - Proceedings - 2017 IEEE International Conference on Computer Vision Workshops, ICCVW 2017
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 16th IEEE International Conference on Computer Vision Workshops, ICCVW 2017
Y2 - 22 October 2017 through 29 October 2017
ER -