Extremely Simple Activation Shaping for Out-of-Distribution Detection
TL;DR
At inference time, pick a layer, simplify its representation, feed it through the rest of the network. Accuracy is not affected and OOD detection is much better!
Venue
In Eleventh International Conference on Learning Representations (ICLR 2020).
BibTeX
@article{djurisic2022ash,
title={Extremely Simple Activation Shaping for Out-of-Distribution Detection},
author={Djurisic, Andrija and Bozanic, Nebojsa and Ashok, Arjun and Liu, Rosanne},
journal={arXiv preprint arXiv:2209.09858},
year={2022}
}
title={Extremely Simple Activation Shaping for Out-of-Distribution Detection},
author={Djurisic, Andrija and Bozanic, Nebojsa and Ashok, Arjun and Liu, Rosanne},
journal={arXiv preprint arXiv:2209.09858},
year={2022}
}
Date
September, 2022