English
Related papers

Related papers: (Almost) Provable Error Bounds Under Distribution …

200 papers

While deep neural networks are highly performant and successful in a wide range of real-world problems, estimating their predictive uncertainty remains a challenging task. To address this challenge, we propose and implement a loss function…

Machine Learning · Computer Science 2022-10-14 Tony Tohme , Kevin Vanslette , Kamal Youcef-Toumi

In statistical classification and machine learning, classification error is an important performance measure, which is minimized by the Bayes decision rule. In practice, the unknown true distribution is usually replaced with a model…

Machine Learning · Computer Science 2025-01-28 Zijian Yang , Vahe Eminyan , Ralf Schlüter , Hermann Ney

This paper considers the generation of prediction intervals (PIs) by neural networks for quantifying uncertainty in regression tasks. It is axiomatic that high-quality PIs should be as narrow as possible, whilst capturing a specified…

Machine Learning · Statistics 2019-04-10 Tim Pearce , Mohamed Zaki , Alexandra Brintrup , Andy Neely

Predictive models that generalize well under distributional shift are often desirable and sometimes crucial to building robust and reliable machine learning applications. We focus on distributional shift that arises in causal inference from…

Machine Learning · Statistics 2018-02-27 Fredrik D. Johansson , Nathan Kallus , Uri Shalit , David Sontag

Self-training often falls short under distribution shifts due to an increased discrepancy between prediction confidence and actual accuracy. This typically necessitates computationally demanding methods such as neighborhood or…

Machine Learning · Computer Science 2024-11-04 Taejong Joo , Diego Klabjan

Conventional wisdom suggests that neural network predictions tend to be unpredictable and overconfident when faced with out-of-distribution (OOD) inputs. Our work reassesses this assumption for neural networks with high-dimensional inputs.…

Machine Learning · Computer Science 2024-03-19 Katie Kang , Amrith Setlur , Claire Tomlin , Sergey Levine

We consider the problem of estimating the class prior in an unlabeled dataset. Under the assumption that an additional labeled dataset is available, the class prior can be estimated by fitting a mixture of class-wise data distributions to…

Machine Learning · Computer Science 2016-11-08 Marthinus C. du Plessis , Gang Niu , Masashi Sugiyama

Covariate shift, a widely used assumption in tackling {\it distributional shift} (when training and test distributions differ), focuses on scenarios where the distribution of the labels conditioned on the feature vector is the same, but the…

Machine Learning · Computer Science 2025-02-24 Deeksha Adil , Jarosław Błasiok

In spite of the dominant performances of deep neural networks, recent works have shown that they are poorly calibrated, resulting in over-confident predictions. Miscalibration can be exacerbated by overfitting due to the minimization of the…

Computer Vision and Pattern Recognition · Computer Science 2023-07-06 Bingyuan Liu , Ismail Ben Ayed , Adrian Galdran , Jose Dolz

The recent decade has seen an enormous rise in the popularity of deep learning and neural networks. These algorithms have broken many previous records and achieved remarkable results. Their outstanding performance has significantly sped up…

Estimating the uncertainty in deep neural network predictions is crucial for many real-world applications. A common approach to model uncertainty is to choose a parametric distribution and fit the data to it using maximum likelihood…

Machine Learning · Computer Science 2022-11-28 Ali Harakeh , Jordan Hu , Naiqing Guan , Steven L. Waslander , Liam Paull

To assess generalization, machine learning scientists typically either (i) bound the generalization gap and then (after training) plug in the empirical risk to obtain a bound on the true risk; or (ii) validate empirically on holdout data.…

Machine Learning · Computer Science 2021-11-09 Saurabh Garg , Sivaraman Balakrishnan , J. Zico Kolter , Zachary C. Lipton

Evaluating the statistical dimension is a common tool to determine the asymptotic phase transition in compressed sensing problems with Gaussian ensemble. Unfortunately, the exact evaluation of the statistical dimension is very difficult and…

Information Theory · Computer Science 2019-06-06 Sajad Daei , Farzan Haddadi , Arash Amini , Martin Lotz

This article studies the achievable guarantees on the error rates of certain learning algorithms, with particular focus on refining logarithmic factors. Many of the results are based on a general technique for obtaining bounds on the error…

Machine Learning · Computer Science 2016-09-13 Steve Hanneke

Given a deep neural network image classification model that we treat as a black box, and an unlabeled evaluation dataset, we develop an efficient strategy by which the classifier can be evaluated. Randomly sampling and labeling instances…

Machine Learning · Computer Science 2020-06-30 Walter Bennette , Karsten Maurer , Sean Sisti

This paper addresses challenges in robust transfer learning stemming from ambiguity in Bayes classifiers and weak transferable signals between the target and source distribution. We introduce a novel quantity called the ''ambiguity level''…

Machine Learning · Statistics 2025-05-06 Jianqing Fan , Cheng Gao , Jason M. Klusowski

This paper presents a conformal prediction method for classification in highly imbalanced and open-set settings, where there are many possible classes and not all may be represented in the data. Existing approaches require a finite, known…

Machine Learning · Statistics 2025-10-16 Tianmin Xie , Yanfei Zhou , Ziyi Liang , Stefano Favaro , Matteo Sesia

We present novel bounds for estimating discrete probability distributions under the $\ell_\infty$ norm. These are nearly optimal in various precise senses, including a kind of instance-optimality. Our data-dependent convergence guarantees…

Statistics Theory · Mathematics 2024-02-14 Aryeh Kontorovich , Amichai Painsky

Mislabeled samples are ubiquitous in real-world datasets as rule-based or expert labeling is usually based on incorrect assumptions or subject to biased opinions. Neural networks can "memorize" these mislabeled samples and, as a result,…

Machine Learning · Computer Science 2021-11-24 Katharina Rombach , Gabriel Michau , Olga Fink

An assumption often made in supervised learning is that the training and testing sets have the same label distribution. However, in real-life scenarios, this assumption rarely holds. For example, medical diagnosis result distributions…

Machine Learning · Computer Science 2026-04-03 Yunrui Zhang , Gustavo Batista , Salil S. Kanhere