Timezone: »

 
Spotlight
Mitigating Neural Network Overconfidence with Logit Normalization
Hongxin Wei · RENCHUNZI XIE · Hao Cheng · LEI FENG · Bo An · Sharon Li

Wed Jul 20 07:55 AM -- 08:00 AM (PDT) @ Room 318 - 320
in Social Aspects/MISC »

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.

Keywords: MISC: Representation Learning · SA: Trustworthy Machine Learning · SA: Fairness, Equity, Justice and Safety

Author Information

Hongxin Wei (Nanyang Technological University)
RENCHUNZI XIE (Nanyang Technological University)
Hao Cheng (Nanjing University)
LEI FENG (Nanyang Technological University)
Bo An (Nanyang Technological University)
Sharon Li (University of Wisconsin-Madison)

Related Events (a corresponding poster, oral, or spotlight)

More from the Same Authors