snap-stanford / crust

[NeurIPS 2020] Coresets for Robust Training of Neural Networks against Noisy Labels

Coresets for Robust Training of Neural Networks against Noisy Labels

Baharan Mirzasoleiman, Kaidi Cao, Jure Leskovec

---

This is the official implementation of crust in the paper Coresets for Robust Training of Neural Networks against Noisy Labels in PyTorch.

Dependency

The code is built with following libraries:

• PyTorch 1.7
• scikit-learn

Training

We provide a training example with this repo:

python robust_cifar_train.py --gpu 0 --use_crust

Reference

If you find our paper and repo useful, please cite as

@article{mirzasoleiman2020coresets,
  title={Coresets for Robust Training of Neural Networks against Noisy Labels},
  author={Mirzasoleiman, Baharan and Cao, Kaidi and Leskovec, Jure},
  journal={Advances in Neural Information Processing Systems},
  volume={33},
  year={2020}
}