中文
相关论文

相关论文: Beyond calibration: estimating the grouping loss o…

200 篇论文

Accurate estimation of predictive uncertainty (model calibration) is essential for the safe application of neural networks. Many instances of miscalibration in modern neural networks have been reported, suggesting a trend that newer, more…

With model trustworthiness being crucial for sensitive real-world applications, practitioners are putting more and more focus on improving the uncertainty calibration of deep neural networks. Calibration errors are designed to quantify the…

机器学习 · 计算机科学 2024-03-14 Sebastian G. Gruber , Florian Buettner

Proper confidence calibration of deep neural networks is essential for reliable predictions in safety-critical tasks. Miscalibration can lead to model over-confidence and/or under-confidence; i.e., the model's confidence in its prediction…

机器学习 · 计算机科学 2023-08-08 Shuang Ao , Stefan Rueger , Advaith Siddharthan

Confidence calibration -- the problem of predicting probability estimates representative of the true correctness likelihood -- is important for classification models in many applications. We discover that modern neural networks, unlike…

机器学习 · 计算机科学 2017-08-04 Chuan Guo , Geoff Pleiss , Yu Sun , Kilian Q. Weinberger

Miscalibration - a mismatch between a model's confidence and its correctness - of Deep Neural Networks (DNNs) makes their predictions hard to rely on. Ideally, we want networks to be accurate, calibrated and confident. We show that, as…

Neural networks solving real-world problems are often required not only to make accurate predictions but also to provide a confidence level in the forecast. The calibration of a model indicates how close the estimated confidence is to the…

神经与进化计算 · 计算机科学 2023-03-21 Ruslan Vasilev , Alexander D'yakonov

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. However, these works have primarily focused on classification accuracy,…

计算机视觉与模式识别 · 计算机科学 2020-09-10 Gongbo Liang , Yu Zhang , Xiaoqin Wang , Nathan Jacobs

Reliable confidence estimation is a challenging yet fundamental requirement in many risk-sensitive applications. However, modern deep neural networks are often overconfident for their incorrect predictions, i.e., misclassified samples from…

计算机视觉与模式识别 · 计算机科学 2024-03-06 Fei Zhu , Xu-Yao Zhang , Zhen Cheng , Cheng-Lin Liu

When probabilistic classifiers are trained and calibrated, the so-called grouping loss component of the calibration loss can easily be overlooked. Grouping loss refers to the gap between observable information and information actually…

机器学习 · 统计学 2022-04-26 Dirk Tasche

Model calibration is essential for ensuring that the predictions of deep neural networks accurately reflect true probabilities in real-world classification tasks. However, deep networks often produce over-confident or under-confident…

机器学习 · 计算机科学 2025-04-01 Jinxu Lin , Linwei Tao , Minjing Dong , Chang Xu

Probabilistic classifiers are central for making informed decisions under uncertainty. Based on the maximum expected utility principle, optimal decision rules can be derived using the posterior class probabilities and misclassification…

机器学习 · 计算机科学 2025-03-25 Alexandre Perez-Lebel , Gael Varoquaux , Sanmi Koyejo , Matthieu Doutreligne , Marine Le Morvan

Calibration is a fundamental property of a good predictive model: it requires that the model predicts correctly in proportion to its confidence. Modern neural networks, however, provide no strong guarantees on their calibration -- and can…

机器学习 · 计算机科学 2022-10-07 A. Michael Carrell , Neil Mallinar , James Lucas , Preetum Nakkiran

For classification models based on neural networks, the maximum predicted class probability is often used as a confidence score. This score rarely predicts well the probability of making a correct prediction and requires a post-processing…

机器学习 · 计算机科学 2024-11-07 Adrien LeCoz , Stéphane Herbin , Faouzi Adjed

Deep neural networks (DNNs) have enabled astounding progress in several vision-based problems. Despite showing high predictive accuracy, recently, several works have revealed that they tend to provide overconfident predictions and thus are…

计算机视觉与模式识别 · 计算机科学 2023-03-28 Muhammad Akhtar Munir , Muhammad Haris Khan , Salman Khan , Fahad Shahbaz Khan

Many applications of classification methods not only require high accuracy but also reliable estimation of predictive uncertainty. However, while many current classification frameworks, in particular deep neural networks, achieve high…

机器学习 · 计算机科学 2020-02-28 Jonathan Wenger , Hedvig Kjellström , Rudolph Triebel

Reliable confidence estimation for the predictions is important in many safety-critical applications. However, modern deep neural networks are often overconfident for their incorrect predictions. Recently, many calibration methods have been…

机器学习 · 计算机科学 2023-03-07 Fei Zhu , Zhen Cheng , Xu-Yao Zhang , Cheng-Lin Liu

Classifier calibration does not always go hand in hand with the classifier's ability to separate the classes. There are applications where good classifier calibration, i.e. the ability to produce accurate probability estimates, is more…

机器学习 · 计算机科学 2020-05-26 Tuomo Alasalmi , Jaakko Suutala , Heli Koskimäki , Juha Röning

Machine learning classifiers often produce probabilistic predictions that are critical for accurate and interpretable decision-making in various domains. The quality of these predictions is generally evaluated with proper losses, such as…

机器学习 · 计算机科学 2025-06-26 Eugène Berta , David Holzmüller , Michael I. Jordan , Francis Bach

Many classification applications require accurate probability estimates in addition to good class separation but often classifiers are designed focusing only on the latter. Calibration is the process of improving probability estimates by…

机器学习 · 计算机科学 2020-01-31 Tuomo Alasalmi , Jaakko Suutala , Heli Koskimäki , Juha Röning

Probabilistic classifiers output a probability distribution on target classes rather than just a class prediction. Besides providing a clear separation of prediction and decision making, the main advantage of probabilistic models is their…

机器学习 · 计算机科学 2019-02-20 Juozas Vaicenavicius , David Widmann , Carl Andersson , Fredrik Lindsten , Jacob Roll , Thomas B. Schön
‹ 上一页 1 2 3 10 下一页 ›