English
Related papers

Related papers: Estimating calibration error under label shift wit…

200 papers

Recalibrating probabilistic classifiers is vital for enhancing the reliability and accuracy of predictive models. Despite the development of numerous recalibration algorithms, there is still a lack of a comprehensive theory that integrates…

Machine Learning · Computer Science 2023-05-19 Zeyu Sun , Dogyoon Song , Alfred Hero

In many real-world applications, researchers aim to deploy models trained in a source domain to a target domain, where obtaining labeled data is often expensive, time-consuming, or even infeasible. While most existing literature assumes…

Methodology · Statistics 2025-08-26 Seong-ho Lee , Yanyuan Ma , Jiwei Zhao

Research has shown that deep networks tend to be overly optimistic about their predictions, leading to an underestimation of prediction errors. Due to the limited nature of data, existing studies have proposed various methods based on model…

Machine Learning · Computer Science 2023-10-24 Jia-Qi Yang , De-Chuan Zhan , Le Gan

Ensuring that predicted probabilities align with observed frequencies is critical in high-stakes domains such as clinical decision support, autonomous driving and financial risk assessment. Existing calibration methods typically apply a…

Machine Learning · Computer Science 2026-05-26 Tomer Lavi , Bracha Shapira , Nadav Rappoport

Calibration requires predictor outputs to be consistent with their Bayesian posteriors. For machine learning predictors that do not distinguish between small perturbations, calibration errors are continuous in predictions, e.g., smooth…

Machine Learning · Computer Science 2025-04-23 Jason Hartline , Yifan Wu , Yunran Yang

The state-of-the-art performance on entity resolution (ER) has been achieved by deep learning. However, deep models are usually trained on large quantities of accurately labeled training data, and can not be easily tuned towards a target…

Machine Learning · Computer Science 2022-04-12 Zhaoqiang Chen , Qun Chen , Youcef Nafa , Tianyi Duan , Wei Pan , Lijun Zhang , Zhanhuai Li

Open set label shift (OSLS) occurs when label distributions change from a source to a target distribution, and the target distribution has an additional out-of-distribution (OOD) class. In this work, we build estimators for both source and…

Machine Learning · Computer Science 2025-05-12 Changkun Ye , Russell Tsuchida , Lars Petersson , Nick Barnes

We propose Regularized Learning under Label shifts (RLLS), a principled and a practical domain-adaptation algorithm to correct for shifts in the label distribution between a source and a target domain. We first estimate importance weights…

Machine Learning · Computer Science 2020-08-10 Kamyar Azizzadenesheli , Anqi Liu , Fanny Yang , Animashree Anandkumar

We consider a semi-supervised classification problem with non-stationary label-shift in which we observe a labelled data set followed by a sequence of unlabelled covariate vectors in which the marginal probabilities of the class labels may…

Statistics Theory · Mathematics 2024-05-29 Henry W J Reeve

Let P represent the source population with complete data, containing covariate $\mathbf{Z}$ and response $T$, and Q the target population, where only the covariate $\mathbf{Z}$ is available. We consider a setting with both label shift and…

Methodology · Statistics 2025-06-27 Yuxiang Zong , Yanyuan Ma , Ingrid Van Keilegom

Instance-dependent label noise is realistic but rather challenging, where the label-corruption process depends on instances directly. It causes a severe distribution shift between the distributions of training and test data, which impairs…

Machine Learning · Computer Science 2022-10-12 Manyi Zhang , Yuxin Ren , Zihao Wang , Chun Yuan

Datasets often contain biases which unfairly disadvantage certain groups, and classifiers trained on such datasets can inherit these biases. In this paper, we provide a mathematical formulation of how this bias can arise. We do so by…

Machine Learning · Computer Science 2019-01-16 Heinrich Jiang , Ofir Nachum

Optimal decision making requires that classifiers produce uncertainty estimates consistent with their empirical accuracy. However, deep neural networks are often under- or over-confident in their predictions. Consequently, methods have been…

Popular zero-shot models suffer due to artifacts inherited from pretraining. One particularly detrimental issue, caused by unbalanced web-scale pretraining data, is mismatched label distribution. Existing approaches that seek to repair the…

Machine Learning · Computer Science 2024-10-31 Changho Shin , Jitian Zhao , Sonia Cromp , Harit Vishwakarma , Frederic Sala

Calibrated predictions are useful because their numerical values can be interpreted as probabilities. Calibration errors are therefore widely used to evaluate, compare, and tune probabilistic predictors. Recently, Haghtalab et al. (2024)…

Machine Learning · Computer Science 2026-05-19 Yuxuan Lu , Yifan Wu , Jason Hartline , Lunjia Hu

A probabilistic model is said to be calibrated if its predicted probabilities match the corresponding empirical frequencies. Calibration is important for uncertainty quantification and decision making in safety-critical applications. While…

Machine Learning · Computer Science 2020-07-01 Anusri Pampari , Stefano Ermon

The Expected Normalized Calibration Error (ENCE) is a popular calibration statistic used in Machine Learning to assess the quality of prediction uncertainties for regression problems. Estimation of the ENCE is based on the binning of…

Machine Learning · Computer Science 2023-05-23 Pascal Pernot

How many labeled examples are needed to estimate a classifier's performance on a new dataset? We study the case where data is plentiful, but labels are expensive. We show that by making a few reasonable assumptions on the structure of the…

Machine Learning · Computer Science 2012-10-09 Peter Welinder , Max Welling , Pietro Perona

Domain Adaptation (DA) enables transferring a learning machine from a labeled source domain to an unlabeled target one. While remarkable advances have been made, most of the existing DA methods focus on improving the target accuracy at…

Machine Learning · Computer Science 2020-11-10 Ximei Wang , Mingsheng Long , Jianmin Wang , Michael I. Jordan

Most machine learning classifiers are designed to output posterior probabilities for the classes given the input sample. These probabilities may be used to make the categorical decision on the class of the sample; provided as input to a…

Machine Learning · Statistics 2024-08-07 Luciana Ferrer , Daniel Ramos