Patch-wise Adversarial Regularization (for Replication Usage)

For Replication Purpose

• This repository contains the code that is used for replicating our reported numbers, so some codes might be messy.
• For clearer demonstration of our method, please visit: HaohanWang/PAR

Reference

If you find this method helpful, please consider cite:

  Haohan Wang, Songwei Ge, Eric P. Xing, and Zachary C. Lipton. "Learning Robust Global Representations by Penalizing Local Predictive Power" (NeurIPS 2019)

Other Resources

• You may be also interested in HaohanWang/HEX_experiments