English
Related papers

Related papers: Can a Confident Prior Replace a Cold Posterior?

200 papers

The non-uniform surface temperature distribution of rotating active stars is routinely mapped with the Doppler Imaging technique. Inhomogeneities in the surface produce features in high-resolution spectroscopic observations that shift in…

Instrumentation and Methods for Astrophysics · Physics 2022-02-23 A. Asensio Ramos , C. Diaz Baso , O. Kochukhov

Neural networks make accurate predictions but often fail to provide reliable uncertainty estimates, especially under covariate distribution shifts between training and testing. To address this problem, we propose a Bayesian framework for…

Machine Learning · Statistics 2025-12-22 Yuli Slavutsky , David M. Blei

Test-time adaptation with pre-trained vision-language models, such as CLIP, aims to adapt the model to new, potentially out-of-distribution test data. Existing methods calculate the similarity between visual embedding and learnable class…

Computer Vision and Pattern Recognition · Computer Science 2025-03-18 Lihua Zhou , Mao Ye , Shuaifeng Li , Nianxin Li , Xiatian Zhu , Lei Deng , Hongbin Liu , Zhen Lei

Prediction sets capture uncertainty by predicting sets of labels rather than individual labels, enabling downstream decisions to conservatively account for all plausible outcomes. Conformal inference algorithms construct prediction sets…

Machine Learning · Statistics 2023-10-20 Wenwen Si , Sangdon Park , Insup Lee , Edgar Dobriban , Osbert Bastani

When faced with distribution shift at test time, deep neural networks often make inaccurate predictions with unreliable uncertainty estimates. While improving the robustness of neural networks is one promising approach to mitigate this…

Machine Learning · Computer Science 2021-09-28 Aurick Zhou , Sergey Levine

Approximate Bayesian inference for neural networks is considered a robust alternative to standard training, often providing good performance on out-of-distribution data. However, Bayesian neural networks (BNNs) with high-fidelity…

Machine Learning · Computer Science 2021-12-07 Pavel Izmailov , Patrick Nicholson , Sanae Lotfi , Andrew Gordon Wilson

It can be important in Bayesian analyses of complex models to construct informative prior distributions which reflect knowledge external to the data at hand. Nevertheless, how much prior information an analyst can elicit from an expert will…

Applications · Statistics 2017-11-10 Xueou Wang , David J. Nott , C. C. Drovandi , Kerrie Mengersen , Michael Evans

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 probability outputs of trained classifiers are increasingly used as inputs to downstream regression estimands such as effects, prevalences, or disparities for a latent group observed only on a small labelled subset. A standard…

Methodology · Statistics 2026-05-14 Marcell T. Kurbucz

A common approach in positive-unlabeled learning is to train a classification model between labeled and unlabeled data. This strategy is in fact known to give an optimal classifier under mild conditions; however, it results in biased…

Machine Learning · Statistics 2017-02-03 Shantanu Jain , Martha White , Predrag Radivojac

Posterior tempering reduces the influence of the likelihood in the calculation of the posterior by raising the likelihood to a fractional power $\alpha$. The resulting power posterior - also known as an $\alpha$-posterior or fractional…

Statistics Theory · Mathematics 2026-01-15 Ruchira Ray , Marco Avella Medina , Cynthia Rush

Bayesian decision theory provides an elegant framework for acting optimally under uncertainty when tractable posterior distributions are available. Modern Bayesian models, however, typically involve intractable posteriors that are…

Machine Learning · Computer Science 2021-06-15 Meet P. Vadera , Soumya Ghosh , Kenney Ng , Benjamin M. Marlin

In image classification tasks, deep learning models are vulnerable to image distortions i.e. their accuracy significantly drops if the input images are distorted. An image-classifier is considered "reliable" if its accuracy on distorted…

Computer Vision and Pattern Recognition · Computer Science 2025-07-23 Dang Nguyen , Sunil Gupta , Kien Do , Svetha Venkatesh

Mixture models and topic models generate each observation from a single cluster, but standard variational posteriors for each observation assign positive probability to all possible clusters. This requires dense storage and runtime costs…

Machine Learning · Statistics 2017-11-15 Michael C. Hughes , Erik B. Sudderth

The information bottleneck framework provides a systematic approach to learning representations that compress nuisance information in the input and extract semantically meaningful information about predictions. However, the choice of a…

Prior distributions for high-dimensional linear regression require specifying a joint distribution for the unobserved regression coefficients, which is inherently difficult. We instead propose a new class of shrinkage priors for linear…

Methodology · Statistics 2020-07-09 Yan Dora Zhang , Brian P. Naughton , Howard D. Bondell , Brian J. Reich

This paper presents a comprehensive empirical analysis of conformal prediction methods on a challenging aerial image dataset featuring diverse events in unconstrained environments. Conformal prediction is a powerful post-hoc technique that…

Machine Learning · Computer Science 2025-04-25 Farhad Pourkamali-Anaraki

Label bias occurs when the outcome of interest is not directly observable and instead, modeling is performed with proxy labels. When the difference between the true outcome and the proxy label is correlated with predictors, this can yield…

Methodology · Statistics 2025-12-02 Jonas Mikhaeil , Andrew Gelman , Philip Greengard

Dataset shift is common in credit scoring scenarios, and the inconsistency between the distribution of training data and the data that actually needs to be predicted is likely to cause poor model performance. However, most of the current…

Machine Learning · Computer Science 2021-12-21 Hongyi Qian , Baohui Wang , Ping Ma , Lei Peng , Songfeng Gao , You Song

Contrastively trained encoders have recently been proven to invert the data-generating process: they encode each input, e.g., an image, into the true latent vector that generated the image (Zimmermann et al., 2021). However, real-world…

Machine Learning · Computer Science 2023-05-18 Michael Kirchhof , Enkelejda Kasneci , Seong Joon Oh