Neural Clamping: Joint Input Perturbation and Temperature Scaling for Neural Network Calibration

by   Yung-Chen Tang, et al.
National Tsing Hua University

Neural network calibration is an essential task in deep learning to ensure consistency between the confidence of model prediction and the true correctness likelihood. In this paper, we propose a new post-processing calibration method called Neural Clamping, which employs a simple joint input-output transformation on a pre-trained classifier via a learnable universal input perturbation and an output temperature scaling parameter. Moreover, we provide theoretical explanations on why Neural Clamping is provably better than temperature scaling. Evaluated on CIFAR-100 and ImageNet image recognition datasets and a variety of deep neural network models, our empirical results show that Neural Clamping significantly outperforms state-of-the-art post-processing calibration methods.


Confidence Calibration with Bounded Error Using Transformations

As machine learning techniques become widely adopted in new domains, esp...

Layer-Stack Temperature Scaling

Recent works demonstrate that early layers in a neural network contain u...

Expectation consistency for calibration of neural networks

Despite their incredible performance, it is well reported that deep neur...

Unsupervised Temperature Scaling: Post-Processing Unsupervised Calibration of Deep Models Decisions

Great performances of deep learning are undeniable, with impressive resu...

Verified Uncertainty Calibration

Applications such as weather forecasting and personalized medicine deman...

Local Temperature Scaling for Probability Calibration

For semantic segmentation, label probabilities are often uncalibrated as...

Please sign up or login with your details

Forgot password? Click here to reset