Post-hoc Uncertainty Learning Using a Dirichlet Meta-Model
Proceedings of the AAAI Conference on Artificial Intelligence · 26.06.2023 · 37 (8) · 9772-9781
DOI: 10.1609/aaai.v37i8.26167 ↗Kurz gefasst
Fachartikel in der Zeitschrift Proceedings of the AAAI Conference on Artificial Intelligence, online erschienen am 26.06.2023. Verfasst von Maohao Shen und 5 weiteren Autor:innen. Die Arbeit wurde bisher 24-mal zitiert (Crossref).
Abstract (Originalsprache)
It is known that neural networks have the problem of being over-confident when directly using the output label distribution to generate uncertainty measures. Existing methods mainly resolve this issue by retraining the entire model to impose the uncertainty quantification capability so that the learned model can achieve desired performance in accuracy and uncertainty prediction simultaneously. However, training the model from scratch is computationally expensive, and a trade-off might exist between prediction accuracy and uncertainty quantification. To this end, we consider a more practical post-hoc uncertainty learning setting, where a well-trained base model is given, and we focus on the uncertainty quantification task at the second stage of training. We propose a novel Bayesian uncertainty learning approach using the Dirichlet meta-model, which is effective and computationally efficient. Our proposed method requires no additional training data and is flexible enough to quantify different uncertainties and easily adapt to different application settings, including out-of-domain data detection, misclassification detection, and trustworthy transfer learning. Finally, we demonstrate our proposed meta-model approach's flexibility and superior empirical performance on these applications over multiple representative image classification benchmarks.
Zitieren
APA
Shen, M., Bu, Y., Sattigeri, P., Ghosh, S., Das, S., & Wornell, G. (2023). Post-hoc Uncertainty Learning Using a Dirichlet Meta-Model. Proceedings of the AAAI Conference on Artificial Intelligence, 37(8), 9772-9781. https://doi.org/10.1609/aaai.v37i8.26167
Vancouver
Shen M, Bu Y, Sattigeri P, Ghosh S, Das S, Wornell G. Post-hoc Uncertainty Learning Using a Dirichlet Meta-Model. Proceedings of the AAAI Conference on Artificial Intelligence. 2023;37(8):9772-9781. doi:10.1609/aaai.v37i8.26167
BibTeX
@article{Shen2023,
author = {Shen, Maohao and Bu, Yuheng and Sattigeri, Prasanna and Ghosh, Soumya and Das, Subhro and Wornell, Gregory},
title = {Post-hoc Uncertainty Learning Using a Dirichlet Meta-Model},
journal = {Proceedings of the AAAI Conference on Artificial Intelligence},
year = {2023},
volume = {37},
number = {8},
pages = {9772--9781},
doi = {10.1609/aaai.v37i8.26167}
}
EUVIMED