A tutorial on calibration measurements and calibration models for clinical prediction models

Y Huang, W Li, F Macheret, RA Gabriel… - Journal of the …, 2020 - academic.oup.com
Our primary objective is to provide the clinical informatics community with an introductory
tutorial on calibration measurements and calibration models for predictive models using …

Revisiting the calibration of modern neural networks

M Minderer, J Djolonga, R Romijnders… - Advances in …, 2021 - proceedings.neurips.cc
Accurate estimation of predictive uncertainty (model calibration) is essential for the safe
application of neural networks. Many instances of miscalibration in modern neural networks …

How can we know what language models know?

Z Jiang, FF Xu, J Araki, G Neubig - Transactions of the Association for …, 2020 - direct.mit.edu
Recent work has presented intriguing results examining the knowledge contained in
language models (LMs) by having the LM fill in the blanks of prompts such as “Obama is a …

How Can We Know When Language Models Know? On the Calibration of Language Models for Question Answering

Z Jiang, J Araki, H Ding, G Neubig - Transactions of the Association …, 2021 - direct.mit.edu
Recent works have shown that language models (LM) capture different types of knowledge
regarding facts or common sense. However, because no model is perfect, they still fail to …

Improving calibration for long-tailed recognition

Z Zhong, J Cui, S Liu, J Jia - … of the IEEE/CVF conference on …, 2021 - openaccess.thecvf.com
Deep neural networks may perform poorly when training datasets are heavily class-
imbalanced. Recently, two-stage methods decouple representation learning and classifier …

Accurate uncertainties for deep learning using calibrated regression

V Kuleshov, N Fenner, S Ermon - … conference on machine …, 2018 - proceedings.mlr.press
Methods for reasoning under uncertainty are a key building block of accurate and reliable
machine learning systems. Bayesian methods provide a general framework to quantify …

Verified uncertainty calibration

A Kumar, PS Liang, T Ma - Advances in Neural Information …, 2019 - proceedings.neurips.cc
Applications such as weather forecasting and personalized medicine demand models that
output calibrated probability estimates---those representative of the true likelihood of a …

[PDF][PDF] Measuring Calibration in Deep Learning.

J Nixon, MW Dusenberry, L Zhang… - CVPR …, 2019 - openaccess.thecvf.com
The reliability of a machine learning model's confidence in its predictions is critical for high-
risk applications. Calibration—the idea that a model's predicted probabilities of outcomes …

On calibration of modern neural networks

C Guo, G Pleiss, Y Sun… - … conference on machine …, 2017 - proceedings.mlr.press
Confidence calibration–the problem of predicting probability estimates representative of the
true correctness likelihood–is important for classification models in many applications. We …

Deep k-nearest neighbors: Towards confident, interpretable and robust deep learning

N Papernot, P McDaniel - arXiv preprint arXiv:1803.04765, 2018 - arxiv.org
Deep neural networks (DNNs) enable innovative applications of machine learning like
image recognition, machine translation, or malware detection. However, deep learning is …