Skip to content
/ PLC Public

ICLR 2021, "Learning with feature-dependent label noise: a progressive approach"

Notifications You must be signed in to change notification settings

pxiangwu/PLC

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

6 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Learning with Feature-Dependent Label Noise: A Progressive Approach (ICLR 2021, spotlight) Paper

Requirements

  • PyTorch 1.6 (Other versions >= 1.0 should also work)
  • Python 3.8.5 (Other Python 3.x should also work)
  • tqdm, termcolor, etc (which can be easily installed via pip)

Usage

  • The folder cifar contains the code for generating PMD noise and running on synthetic datasets.
  • The folder clothing1m contains the code for running on Clothing1M dataset.
  • The folder Food101 contains the code for running on Food-101N dataset.

Reference

@inproceedings{prog_noise_iclr2021,
  title={Learning with Feature-Dependent Label Noise: A Progressive Approach},
  author={Zhang, Yikai and Zheng, Songzhu and Wu, Pengxiang and Goswami, Mayank and Chen, Chao},
  booktitle={ICLR},
  year={2021}
}

Related Work

  • Error-Bounded Correction of Noisy Labels. In ICML, 2020. [Paper][Code]
  • A Topological Filter for Learning with Label Noise. In NeurIPS, 2020. [Paper][Code]

About

ICLR 2021, "Learning with feature-dependent label noise: a progressive approach"

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages