Importance Weighting for Unlabeled-unlabeled Learning under Distribution Shift
- URL: http://arxiv.org/abs/2609.10994v1
- Date: Thu, 10 Sep 2026 02:13:53 GMT
- Title: Importance Weighting for Unlabeled-unlabeled Learning under Distribution Shift
- Abstract summary: Unlabeled-unlabeled (UU) learning allows us to learn a binary classifier from two sets of unlabeled data with different class-priors.<n>This paper proposes a distribution shift adaptation method for UU learning that uses UU data in the training distribution and a few UU data in the test distribution.<n>Thanks to the generality of UU learning, our method can handle various learning problems such as PU and noisy label learning under distribution shift within a single framework.
- Score: 35.99400750792333
- License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
- Abstract: Unlabeled-unlabeled (UU) learning allows us to learn a binary classifier from two sets of unlabeled data with different class-priors. It is a general framework because it includes a wide variety of supervised learning such as positive-unlabeled (PU) learning, noisy label learning, and similarity-based learning. Existing UU learning assumes that the test and training distributions have the same class-conditional densities. However, this assumption rarely holds in practice due to distribution shifts. This paper proposes a distribution shift adaptation method for UU learning that uses UU data in the training distribution and a few UU data in the test distribution. The proposed method is based on the importance weighting, which minimizes the test risk by using training data with estimated importance weights. Although existing importance weighting methods cannot handle UU data, we show that it can be done in a principled manner. Thanks to the generality of UU learning, our method can handle various learning problems such as PU and noisy label learning under distribution shift within a single framework while existing methods are usually tailored to a specific problem. Moreover, it does not require any assumption of the shift types such as covariate shift. We experimentally demonstrate the effectiveness of the proposed method with real-world datasets.
Related papers
- Improving realistic semi-supervised learning with doubly robust estimation [8.828699635463265]
A major challenge in Semi-Supervised Learning (SSL) is the limited information available about the class distribution in the unlabeled data.<n>We propose to explicitly estimate the unlabeled class distribution, which is a finite-dimensional parameter, emphas an initial step, using a doubly robust estimator with a strong theoretical guarantee.<n>This estimate can then be integrated into existing methods to pseudo-label the unlabeled data during training more accurately.
arXiv Detail & Related papers (2025-02-01T02:34:12Z) - Learning with Complementary Labels Revisited: The Selected-Completely-at-Random Setting Is More Practical [66.57396042747706]
Complementary-label learning is a weakly supervised learning problem.
We propose a consistent approach that does not rely on the uniform distribution assumption.
We find that complementary-label learning can be expressed as a set of negative-unlabeled binary classification problems.
arXiv Detail & Related papers (2023-11-27T02:59:17Z) - Adaptive Negative Evidential Deep Learning for Open-set Semi-supervised Learning [69.81438976273866]
Open-set semi-supervised learning (Open-set SSL) considers a more practical scenario, where unlabeled data and test data contain new categories (outliers) not observed in labeled data (inliers)
We introduce evidential deep learning (EDL) as an outlier detector to quantify different types of uncertainty, and design different uncertainty metrics for self-training and inference.
We propose a novel adaptive negative optimization strategy, making EDL more tailored to the unlabeled dataset containing both inliers and outliers.
arXiv Detail & Related papers (2023-03-21T09:07:15Z) - Dist-PU: Positive-Unlabeled Learning from a Label Distribution
Perspective [89.5370481649529]
We propose a label distribution perspective for PU learning in this paper.
Motivated by this, we propose to pursue the label distribution consistency between predicted and ground-truth label distributions.
Experiments on three benchmark datasets validate the effectiveness of the proposed method.
arXiv Detail & Related papers (2022-12-06T07:38:29Z) - Learning from Positive and Unlabeled Data with Augmented Classes [17.97372291914351]
We propose an unbiased risk estimator for PU learning with Augmented Classes (PUAC)
We derive the estimation error bound for the proposed estimator, which provides a theoretical guarantee for its convergence to the optimal solution.
arXiv Detail & Related papers (2022-07-27T03:40:50Z) - Agree to Disagree: Diversity through Disagreement for Better
Transferability [54.308327969778155]
We propose D-BAT (Diversity-By-disAgreement Training), which enforces agreement among the models on the training data.
We show how D-BAT naturally emerges from the notion of generalized discrepancy.
arXiv Detail & Related papers (2022-02-09T12:03:02Z) - Learning with Proper Partial Labels [87.65718705642819]
Partial-label learning is a kind of weakly-supervised learning with inexact labels.
We show that this proper partial-label learning framework includes many previous partial-label learning settings.
We then derive a unified unbiased estimator of the classification risk.
arXiv Detail & Related papers (2021-12-23T01:37:03Z) - Positive-Unlabeled Classification under Class-Prior Shift: A
Prior-invariant Approach Based on Density Ratio Estimation [85.75352990739154]
We propose a novel PU classification method based on density ratio estimation.
A notable advantage of our proposed method is that it does not require the class-priors in the training phase.
arXiv Detail & Related papers (2021-07-11T13:36:53Z)
This list is automatically generated from the titles and abstracts of the papers in this site.
This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.