Hopfield Boosting for Out-of-Distribution Detection

Abstract

Out-of-distribution (OOD) detection is crucial for real-world machine learning. Outlier exposure methods, which use auxiliary outlier data, can significantly enhance OOD detection. We present Hopfield Boosting, a boosting technique employing modern Hopfield energy (MHE) to refine the boundary between in-distribution (ID) and OOD data. Our method focuses on challenging outlier examples near the decision boundary, achieving a 40% improvement in FPR95 on CIFAR-10, setting a new OOD detection state-of-the-art with outlier exposure.

PDF NeurIPSW OpenReview Semantic Scholar

Cite

Text

Hofmann et al. "Hopfield Boosting for Out-of-Distribution Detection." NeurIPS 2023 Workshops: AMHN, 2023.

Markdown

[Hofmann et al. "Hopfield Boosting for Out-of-Distribution Detection." NeurIPS 2023 Workshops: AMHN, 2023.](https://mlanthology.org/neuripsw/2023/hofmann2023neuripsw-hopfield/)

BibTeX

@inproceedings{hofmann2023neuripsw-hopfield,
  title     = {{Hopfield Boosting for Out-of-Distribution Detection}},
  author    = {Hofmann, Claus and Schmid, Simon Lucas and Lehner, Bernhard and Klotz, Daniel and Hochreiter, Sepp},
  booktitle = {NeurIPS 2023 Workshops: AMHN},
  year      = {2023},
  url       = {https://mlanthology.org/neuripsw/2023/hofmann2023neuripsw-hopfield/}
}