https://github.com/hendrycks/ss-ood

Self-Supervised Learning for OOD Detection (NeurIPS 2019)
https://github.com/hendrycks/ss-ood

ml-safety out-of-distribution-detection robustness self-supervised self-supervised-learning uncertainty

Last synced: 3 months ago
JSON representation

Self-Supervised Learning for OOD Detection (NeurIPS 2019)

• Host: GitHub
• URL: https://github.com/hendrycks/ss-ood
• Owner: hendrycks
• License: mit
• Created: 2019-06-10T02:15:16.000Z (about 6 years ago)
• Default Branch: master
• Last Pushed: 2021-04-29T17:42:13.000Z (about 4 years ago)
• Last Synced: 2025-04-13T04:15:31.468Z (3 months ago)
• Topics: ml-safety, out-of-distribution-detection, robustness, self-supervised, self-supervised-learning, uncertainty
• Language: Python
• Homepage:
• Size: 8.11 MB
• Stars: 266
• Watchers: 7
• Forks: 30
• Open Issues: 3
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# Using Self-Supervised Learning Can Improve Model Robustness and Uncertainty

This repository contains the dataset and some code for the paper [Using Self-Supervised Learning Can Improve Model Robustness and Uncertainty](https://arxiv.org/abs/1906.12340) by Dan Hendrycks, Mantas Mazeika, Saurav Kadavath, and Dawn Song.

We show that self-supervised learning can tremendously improve out-of-distribution detection as well as various types of robustness.



Download the one class ImageNet test set [here](https://drive.google.com/file/d/13xzVuQMEhSnBRZr-YaaO08coLU2dxAUq/view?usp=sharing). The one class ImageNet training set is [here](https://drive.google.com/file/d/1B5c39Fc3haOPzlehzmpTLz6xLtGyKEy4/view?usp=sharing).

The code requires PyTorch 1.0 + and Python 3+.

## Citation

If you find this useful in your research, please consider citing:

    @article{hendrycks2019selfsupervised,

      title={Using Self-Supervised Learning Can Improve Model Robustness and Uncertainty},

      author={Dan Hendrycks and Mantas Mazeika and Saurav Kadavath and Dawn Song},

      journal={Advances in Neural Information Processing Systems (NeurIPS)},

      year={2019}

    }