English
Related papers

Related papers: Tackling Combinatorial Distribution Shift: A Matri…

200 papers

In the face of dataset shift, model calibration plays a pivotal role in ensuring the reliability of machine learning systems. Calibration error (CE) is an indicator of the alignment between the predicted probabilities and the classifier…

Machine Learning · Computer Science 2023-12-15 Teodora Popordanoska , Gorjan Radevski , Tinne Tuytelaars , Matthew B. Blaschko

In a split conformal framework with $K$ classes, a calibration sample of $n$ labeled examples is observed for inference on the label of a new unlabeled example. We explore the setting where a `batch' of $m$ independent such unlabeled…

Methodology · Statistics 2025-03-19 Ulysse Gazin , Ruth Heller , Etienne Roquain , Aldo Solari

In many practical applications of machine learning, a discrepancy often arises between a source distribution from which labeled training examples are drawn and a target distribution for which only unlabeled data is observed. Traditionally,…

Machine Learning · Statistics 2025-03-05 Paweł Teisseyre , Jan Mielniczuk

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

Under label shift, the label distribution p(y) might change but the class-conditional distributions p(x|y) do not. There are two dominant approaches for estimating the label marginal. BBSE, a moment-matching approach based on confusion…

Machine Learning · Computer Science 2020-10-20 Saurabh Garg , Yifan Wu , Sivaraman Balakrishnan , Zachary C. Lipton

When machine learning models are deployed on a test distribution different from the training distribution, they can perform poorly, but overestimate their performance. In this work, we aim to better estimate a model's performance under…

Machine Learning · Computer Science 2020-07-08 Ching-Yao Chuang , Antonio Torralba , Stefanie Jegelka

Training models that perform well under distribution shifts is a central challenge in machine learning. In this paper, we introduce a modeling framework where, in addition to training data, we have partial structural knowledge of the…

Machine Learning · Computer Science 2021-10-28 Tobias Sutter , Andreas Krause , Daniel Kuhn

Rapid progress in representation learning has led to a proliferation of embedding models, and to associated challenges of model selection and practical application. It is non-trivial to assess a model's generalizability to new, candidate…

Machine Learning · Computer Science 2022-02-18 Leo Betthauser , Urszula Chajewska , Maurice Diesendruck , Rohith Pesala

In studies ranging from clinical medicine to policy research, complete data are usually available from a population $\mathscr{P}$, but the quantity of interest is often sought for a related but different population $\mathscr{Q}$ which only…

Methodology · Statistics 2023-07-11 Seong-ho Lee , Yanyuan Ma , Jiwei Zhao

A common assumption in semi-supervised learning is that the labeled, unlabeled, and test data are drawn from the same distribution. However, this assumption is not satisfied in many applications. In many scenarios, the data is collected…

Information Theory · Computer Science 2022-02-25 Gholamali Aminian , Mahed Abroshan , Mohammad Mahdi Khalili , Laura Toni , Miguel R. D. Rodrigues

Machine learning systems may encounter unexpected problems when the data distribution changes in the deployment environment. A major reason is that certain combinations of domains and labels are not observed during training but appear in…

Machine Learning · Computer Science 2022-08-04 Yivan Zhang , Jindong Wang , Xing Xie , Masashi Sugiyama

The vulnerability of models to data aberrations and adversarial attacks influences their ability to demarcate distinct class boundaries efficiently. The network's confidence and uncertainty play a pivotal role in weight adjustments and the…

Machine Learning · Computer Science 2020-12-15 Utkarsh Uppal , Bharat Giddwani

Quantification learning deals with the task of estimating the target label distribution under label shift. In this paper, we first present a unifying framework, distribution feature matching (DFM), that recovers as particular instances…

Machine Learning · Statistics 2023-07-04 Bastien Dussap , Gilles Blanchard , Badr-Eddine Chérief-Abdellatif

Label distribution learning (LDL) is an effective method to predict the relative label description degree (a.k.a. label distribution) of a sample. However, the label distribution is not a complete representation of an instance because it…

Machine Learning · Computer Science 2025-05-29 Jiawei Tang , Yuheng Jia

Test-time adaptation (TTA) aims to adapt a pre-trained model to the target domain in a batch-by-batch manner during inference. While label distributions often exhibit imbalances in real-world scenarios, most previous TTA approaches…

Computer Vision and Pattern Recognition · Computer Science 2023-08-21 Sunghyun Park , Seunghan Yang , Jaegul Choo , Sungrack Yun

As a crucial step toward real-world learning scenarios with changing environments, dataset shift theory and invariant representation learning algorithm have been extensively studied to relax the identical distribution assumption in…

Machine Learning · Computer Science 2024-06-25 You-Wei Luo , Chuan-Xian Ren

As training datasets grow larger, we aspire to develop models that generalize well to any diverse test distribution, even if the latter deviates significantly from the training data. Various approaches like domain adaptation, domain…

Machine Learning · Computer Science 2024-10-10 Andreas Loukas , Karolis Martinkus , Ed Wagstaff , Kyunghyun Cho

In this work, we introduce statistical testing under distributional shifts. We are interested in the hypothesis $P^* \in H_0$ for a target distribution $P^*$, but observe data from a different distribution $Q^*$. We assume that $P^*$ is…

Methodology · Statistics 2022-05-03 Nikolaj Thams , Sorawit Saengkyongam , Niklas Pfister , Jonas Peters

Conformal prediction (CP) offers distribution-free marginal coverage guarantees under an exchangeability assumption, but these guarantees can fail if the data distribution shifts. We analyze the use of pseudo-calibration as a tool to…

Machine Learning · Computer Science 2026-02-17 Farbod Siahkali , Ashwin Verma , Vijay Gupta

In contrast to the standard classification paradigm where the true class is given to each training pattern, complementary-label learning only uses training patterns each equipped with a complementary label, which only specifies one of the…

Machine Learning · Statistics 2019-11-20 Takashi Ishida , Gang Niu , Aditya Krishna Menon , Masashi Sugiyama