Get our free extension to see links to code for papers anywhere online!

Chrome logo  Add to Chrome

Firefox logo Add to Firefox

Boosting Adversarial Robustness From The Perspective of Effective Margin Regularization


Oct 11, 2022
Ziquan Liu, Antoni B. Chan

* BMVC 2022 

   Access Paper or Ask Questions

  • Share via Twitter
  • Share via Facebook
  • Share via LinkedIn
  • Share via Whatsapp
  • Share via Messenger
  • Share via Email

An Empirical Study on Distribution Shift Robustness From the Perspective of Pre-Training and Data Augmentation


May 25, 2022
Ziquan Liu, Yi Xu, Yuanhong Xu, Qi Qian, Hao Li, Rong Jin, Xiangyang Ji, Antoni B. Chan


   Access Paper or Ask Questions

  • Share via Twitter
  • Share via Facebook
  • Share via LinkedIn
  • Share via Whatsapp
  • Share via Messenger
  • Share via Email

Improved Fine-tuning by Leveraging Pre-training Data: Theory and Practice


Nov 24, 2021
Ziquan Liu, Yi Xu, Yuanhong Xu, Qi Qian, Hao Li, Antoni Chan, Rong Jin


   Access Paper or Ask Questions

  • Share via Twitter
  • Share via Facebook
  • Share via LinkedIn
  • Share via Whatsapp
  • Share via Messenger
  • Share via Email

The Implicit Biases of Stochastic Gradient Descent on Deep Neural Networks with Batch Normalization


Feb 06, 2021
Ziquan Liu, Yufei Cui, Jia Wan, Yu Mao, Antoni B. Chan

* Preprint 

   Access Paper or Ask Questions

  • Share via Twitter
  • Share via Facebook
  • Share via LinkedIn
  • Share via Whatsapp
  • Share via Messenger
  • Share via Email

Bayesian Nested Neural Networks for Uncertainty Calibration and Adaptive Compression


Jan 27, 2021
Yufei Cui, Ziquan Liu, Qiao Li, Yu Mao, Antoni B. Chan, Chun Jason Xue

* 16 pages, 10 figures 

   Access Paper or Ask Questions

  • Share via Twitter
  • Share via Facebook
  • Share via LinkedIn
  • Share via Whatsapp
  • Share via Messenger
  • Share via Email

Improve Generalization and Robustness of Neural Networks via Weight Scale Shifting Invariant Regularizations


Aug 07, 2020
Ziquan Liu, Yufei Cui, Antoni B. Chan

* 14 pages, 5 figures 

   Access Paper or Ask Questions

  • Share via Twitter
  • Share via Facebook
  • Share via LinkedIn
  • Share via Whatsapp
  • Share via Messenger
  • Share via Email