Well-calibrated Model Uncertainty with Temperature Scaling for Dropout Variational Inference

Autor:	Laves, Max-Heinrich, Ihler, Sontje, Kortmann, Karl-Philipp, Ortmaier, Tobias
Rok vydání:	2019
Předmět:	Computer Science - Machine Learning Statistics - Machine Learning
Druh dokumentu:	Working Paper
Popis:	Model uncertainty obtained by variational Bayesian inference with Monte Carlo dropout is prone to miscalibration. The uncertainty does not represent the model error well. In this paper, temperature scaling is extended to dropout variational inference to calibrate model uncertainty. Expected uncertainty calibration error (UCE) is presented as a metric to measure miscalibration of uncertainty. The effectiveness of this approach is evaluated on CIFAR-10/100 for recent CNN architectures. Experimental results show, that temperature scaling considerably reduces miscalibration by means of UCE and enables robust rejection of uncertain predictions. The proposed approach can easily be derived from frequentist temperature scaling and yields well-calibrated model uncertainty. It is simple to implement and does not affect the model accuracy. Comment: Accepted at 4th workshop on Bayesian Deep Learning (NeurIPS 2019)
Databáze:	arXiv
Externí odkaz:	http://arxiv.org/abs/1909.13550 Zobrazit plný text záznamu View this record from Arxiv