R Krishnan, O Tickoo - Advances in Neural Information …, 2020 - proceedings.neurips.cc
Obtaining reliable and accurate quantification of uncertainty estimates from deep neural networks is important in safety-critical applications. A well-calibrated model should be …
This paper studies the problem of post-hoc calibration of machine learning classifiers. We introduce the following desiderata for uncertainty calibration:(a) accuracy-preserving,(b) …
Calibrating neural networks is of utmost importance when employing them in safety-critical applications where the downstream decision making depends on the predicted probabilities …
Fine-tuned pre-trained language models can suffer from severe miscalibration for both in- distribution and out-of-distribution (OOD) data due to over-parameterization. To mitigate this …
J Wenger, H Kjellström… - … Conference on Artificial …, 2020 - proceedings.mlr.press
Many applications of classification methods not only require high accuracy but also reliable estimation of predictive uncertainty. However, while many current classification frameworks …
Recent works have shown that deep neural networks can achieve super-human performance in a wide range of image classification tasks in the medical imaging domain …
C Gupta, A Podkopaev… - Advances in Neural …, 2020 - proceedings.neurips.cc
We study three notions of uncertainty quantification---calibration, confidence intervals and prediction sets---for binary classification in the distribution-free setting, that is without making …
Predicting calibrated confidence scores for multi-class deep networks is important for avoiding rare but costly mistakes. A common approach is to learn a post-hoc calibration …
The ability to estimate the uncertainty of predictions made by a neural network is essential when applying neural networks to tasks such as medical diagnosis and autonomous …