Mitigating Neural Network Overconfidence with Logit Normalization
H. Wei, R. Xie, H. Cheng, L. Feng, B. An, и Y. Li. Proceedings of the 39th International Conference on Machine Learning, том 162 из Proceedings of Machine Learning Research, стр. 23631--23644. PMLR, (17--23 Jul 2022)
Аннотация
Detecting out-of-distribution inputs is critical for the safe deployment of machine learning models in the real world. However, neural networks are known to suffer from the overconfidence issue, where they produce abnormally high confidence for both in- and out-of-distribution inputs. In this work, we show that this issue can be mitigated through Logit Normalization (LogitNorm)—a simple fix to the cross-entropy loss—by enforcing a constant vector norm on the logits in training. Our method is motivated by the analysis that the norm of the logit keeps increasing during training, leading to overconfident output. Our key idea behind LogitNorm is thus to decouple the influence of output’s norm during network optimization. Trained with LogitNorm, neural networks produce highly distinguishable confidence scores between in- and out-of-distribution data. Extensive experiments demonstrate the superiority of LogitNorm, reducing the average FPR95 by up to 42.30% on common benchmarks.
%0 Conference Paper
%1 pmlr-v162-wei22d
%A Wei, Hongxin
%A Xie, Renchunzi
%A Cheng, Hao
%A Feng, Lei
%A An, Bo
%A Li, Yixuan
%B Proceedings of the 39th International Conference on Machine Learning
%D 2022
%E Chaudhuri, Kamalika
%E Jegelka, Stefanie
%E Song, Le
%E Szepesvari, Csaba
%E Niu, Gang
%E Sabato, Sivan
%I PMLR
%K LogitNorm OOD
%P 23631--23644
%T Mitigating Neural Network Overconfidence with Logit Normalization
%U https://proceedings.mlr.press/v162/wei22d.html
%V 162
%X Detecting out-of-distribution inputs is critical for the safe deployment of machine learning models in the real world. However, neural networks are known to suffer from the overconfidence issue, where they produce abnormally high confidence for both in- and out-of-distribution inputs. In this work, we show that this issue can be mitigated through Logit Normalization (LogitNorm)—a simple fix to the cross-entropy loss—by enforcing a constant vector norm on the logits in training. Our method is motivated by the analysis that the norm of the logit keeps increasing during training, leading to overconfident output. Our key idea behind LogitNorm is thus to decouple the influence of output’s norm during network optimization. Trained with LogitNorm, neural networks produce highly distinguishable confidence scores between in- and out-of-distribution data. Extensive experiments demonstrate the superiority of LogitNorm, reducing the average FPR95 by up to 42.30% on common benchmarks.
@inproceedings{pmlr-v162-wei22d,
abstract = {Detecting out-of-distribution inputs is critical for the safe deployment of machine learning models in the real world. However, neural networks are known to suffer from the overconfidence issue, where they produce abnormally high confidence for both in- and out-of-distribution inputs. In this work, we show that this issue can be mitigated through Logit Normalization (LogitNorm)—a simple fix to the cross-entropy loss—by enforcing a constant vector norm on the logits in training. Our method is motivated by the analysis that the norm of the logit keeps increasing during training, leading to overconfident output. Our key idea behind LogitNorm is thus to decouple the influence of output’s norm during network optimization. Trained with LogitNorm, neural networks produce highly distinguishable confidence scores between in- and out-of-distribution data. Extensive experiments demonstrate the superiority of LogitNorm, reducing the average FPR95 by up to 42.30% on common benchmarks.},
added-at = {2023-03-06T19:09:25.000+0100},
author = {Wei, Hongxin and Xie, Renchunzi and Cheng, Hao and Feng, Lei and An, Bo and Li, Yixuan},
biburl = {https://www.bibsonomy.org/bibtex/247d55272df04d0eedbfdd15567234716/andolab},
booktitle = {Proceedings of the 39th International Conference on Machine Learning},
editor = {Chaudhuri, Kamalika and Jegelka, Stefanie and Song, Le and Szepesvari, Csaba and Niu, Gang and Sabato, Sivan},
interhash = {7d6a589e9cf59b465345761761c5e08d},
intrahash = {47d55272df04d0eedbfdd15567234716},
keywords = {LogitNorm OOD},
month = {17--23 Jul},
pages = {23631--23644},
pdf = {https://proceedings.mlr.press/v162/wei22d/wei22d.pdf},
publisher = {PMLR},
series = {Proceedings of Machine Learning Research},
timestamp = {2023-03-06T19:09:25.000+0100},
title = {Mitigating Neural Network Overconfidence with Logit Normalization},
url = {https://proceedings.mlr.press/v162/wei22d.html},
volume = 162,
year = 2022
}