Improving Uncertainty Calibration of Deep Neural Networks via Truth Discovery and Geometric Optimization

Chunwei Ma, Ziyun Huang, Jiayi Xian, Mingchen Gao, Jinhui Xu

Research output: Contribution to journalConference articlepeer-review

2 Scopus citations

Abstract

Deep Neural Networks (DNNs), despite their tremendous success in recent years, could still cast doubts on their predictions due to the intrinsic uncertainty associated with their learning process. Ensemble techniques and post-hoc calibrations are two types of approaches that have individually shown promise in improving the uncertainty calibration of DNNs. However, the synergistic effect of the two types of methods has not been well explored. In this paper, we propose a truth discovery framework to integrate ensemble-based and post-hoc calibration methods. Using the geometric variance of the ensemble candidates as a good indicator for sample uncertainty, we design an accuracy-preserving truth estimator with provably no accuracy drop. Furthermore, we show that post-hoc calibration can also be enhanced by truth discovery-regularized optimization. On large-scale datasets including CIFAR and ImageNet, our method shows consistent improvement against state-of-the-art calibration approaches on both histogram-based and kernel density-based evaluation metrics. Our code is available at https://github.com/horsepurve/truly-uncertain.

Original languageEnglish (US)
Pages (from-to)75-85
Number of pages11
JournalProceedings of Machine Learning Research
Volume161
StatePublished - 2021
Event37th Conference on Uncertainty in Artificial Intelligence, UAI 2021 - Virtual, Online
Duration: Jul 27 2021Jul 30 2021

All Science Journal Classification (ASJC) codes

  • Artificial Intelligence
  • Software
  • Control and Systems Engineering
  • Statistics and Probability

Cite this