@inproceedings{843e2a432ef14e6da6c5e77f6418e7f6,
title = "Decoupled Networks",
abstract = "Inner product-based convolution has been a central component of convolutional neural networks (CNNs) and the key to learning visual representations. Inspired by the observation that CNN-learned features are naturally decoupled with the norm of features corresponding to the intra-class variation and the angle corresponding to the semantic difference, we propose a generic decoupled learning framework which models the intra-class variation and semantic difference independently. Specifically, we first reparametrize the inner product to a decoupled form and then generalize it to the decoupled convolution operator which serves as the building block of our decoupled networks. We present several effective instances of the decoupled convolution operator. Each decoupled operator is well motivated and has an intuitive geometric interpretation. Based on these decoupled operators, we further propose to directly learn the operator from data. Extensive experiments show that such decoupled reparameterization renders significant performance gain with easier convergence and stronger robustness.",
author = "Weiyang Liu and Zhen Liu and Zhiding Yu and Bo Dai and Rongmei Lin and Yisen Wang and Rehg, {James M.} and Le Song",
note = "Funding Information: The project was supported in part by NSF IIS- 1218749, NSF Award BCS-1524565 Publisher Copyright: {\textcopyright} 2018 IEEE.; 31st Meeting of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2018 ; Conference date: 18-06-2018 Through 22-06-2018",
year = "2018",
month = dec,
day = "14",
doi = "10.1109/CVPR.2018.00293",
language = "English (US)",
series = "Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition",
publisher = "IEEE Computer Society",
pages = "2771--2779",
booktitle = "Proceedings - 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2018",
address = "United States",
}