/ONI

PyTorch and Torch implementation for our accepted CVPR 2020 paper (Oral): Controllable Orthogonalization in Training DNNs

Primary LanguagePythonBSD 2-Clause "Simplified" LicenseBSD-2-Clause

Orthogonalization by Newton's Iteration (ONI)

Code for reproducing the results of our accpeted CVPR 2020 paper (Oral): Controllable Orthogonalization in Training DNNs (arXiv:2004.00917)

This repo provides the Pytorch and Torch implementations of our methods.

  • ONI_PyTorch: Pytorch project for ImageNet classification and GAN experiments.
  • ONI_Torch: Torch project for Cifar-10 Classfication.

Citations

If you find this repo benefits your research, please consider citing:

@inproceedings{2020_CVPR_Huang,
  author    = {Lei Huang and Li Liu and Fan Zhu and Diwen Wan and Zehuan Yuan and Bo Li and Ling Shao},
  title     = {Controllable Orthogonalization in Training DNNs},
  booktitle = {IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},
  year      = {2020},
  }

Contact

Email: huangleig@nlsde.buaa.edu.cn.