paper-with-me

홈 › Papers

Calibration of Model Uncertainty for Dropout Variational Inference

2020-06-20 · Max-Heinrich Laves, Sontje Ihler, Karl-Philipp Kortmann, Tobias Ortmaier

The model uncertainty obtained by variational Bayesian inference with Monte Carlo dropout is prone to miscalibration. In this paper, different logit scaling methods are extended to dropout variational inference to recalibrate model uncertainty. Expected uncertainty calibration error (UCE) is presented as a metric to measure miscalibration. The effectiveness of recalibration is evaluated on CIFAR-10/100 and SVHN for recent CNN architectures. Experimental results show that logit scaling considerably reduce miscalibration by means of UCE. Well-calibrated uncertainty enables reliable rejection of uncertain predictions and robust detection of out-of-distribution data.

📄 PDF Abstract BibTeX arXiv:2006.11584

Code (0)

등록된 구현이 없습니다.

Tasks

Bayesian InferenceVariational Inference

Methods 이 논문이 사용한 방법론

Monte Carlo Dropout 설명 없음
Dropout Dropout is a regularization technique for neural networks that drops a unit (along with connections) at training time with a specified probability $p$ (a common value is…

Similar Papers 제목 키워드 기반

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

2019-09-30 · Max-Heinrich Laves, Sontje Ihler, Karl-Philipp Kortmann, Tobias Ortmaier

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…

Bayesian InferenceVariational Inference

Robustly representing uncertainty in deep neural networks through sampling

2016-11-05 · Patrick McClure, Nikolaus Kriegeskorte

As deep neural networks (DNNs) are applied to increasingly challenging problems, they will need to be able to represent their own uncertainty. Modeling uncertainty is one of the key features of Bayesian methods. Using Be…

Variational Inference

GFlowOut: Dropout with Generative Flow Networks

2022-10-24 · Dianbo Liu, Moksh Jain, Bonaventure Dossou, Qianli Shen 외

Bayesian Inference offers principled tools to tackle many critical problems with modern neural networks such as poor calibration and generalization, and data inefficiency. However, scaling Bayesian inference to large arc…

Bayesian InferenceVariational Inference

Recalibration of Aleatoric and Epistemic Regression Uncertainty in Medical Imaging

2021-04-26 · Max-Heinrich Laves, Sontje Ihler, Jacob F. Fast, Lüder A. Kahrs 외

The consideration of predictive uncertainty in medical imaging with deep learning is of utmost importance. We apply estimation of both aleatoric and epistemic uncertainty by variational Bayesian inference with Monte Carl…

Bayesian Inferenceregression

Federated Learning via Meta-Variational Dropout

2025-10-23 · Insu Jeon, Minui Hong, Junhyeog Yun, Gunhee Kim arxiv

Federated Learning (FL) aims to train a global inference model from remotely distributed clients, gaining popularity due to its benefit of improving data privacy. However, traditional FL often faces challenges in practic…

Federated Learning