• Home
  • Search
  • Robust Positive-Unlabeled Learning via Bounded Loss Functions under Label Noise
  • https://doi.org/10.64539/sjer.v1i3.2025.314Copy DOI Icon

Robust Positive-Unlabeled Learning via Bounded Loss Functions under Label Noise

Show More
  • Abstract
  • Literature Map
  • Similar Papers
Abstract

Positive-Unlabeled (PU) learning has become a pivotal tool in scenarios where only positive samples are labeled, and negative labels are unavailable. However, in practical applications, the labeled positive data often contains noise such as mislabeled or outlier instances that can severely degrade model performance. This issue is exacerbated using traditional surrogate loss functions, many of which are unbounded and overly sensitive to mislabeled examples. To address this limitation, we propose a robust PU learning framework that integrates bounded loss functions, including ramp loss and truncated logistic loss, into the non-negative risk estimation paradigm. Unlike conventional loss formulations that allow noisy samples to disproportionately influence training, our approach caps each instance’s contribution, thereby reducing the sensitivity to label noise. We mathematically reformulate the PU risk estimator using bounded surrogates and demonstrate that this formulation maintains risk consistency while offering improved noise tolerance. A detailed framework diagram and algorithmic description are provided, along with theoretical analysis that bounds the influence of corrupted labels. Extensive experiments are conducted on both synthetic and real-world datasets under varying noise levels. Our method consistently outperforms baseline models such as unbiased PU (uPU) and non-negative PU (nnPU) in terms of classification accuracy, area under the receiver operating characteristic curve (ROC AUC), and precision-recall area under the curve (PR AUC). The ramp loss variant exhibits particularly strong robustness without sacrificing optimization efficiency. These results demonstrate that incorporating bounded losses is a principled and effective strategy for enhancing the reliability of PU learning in noisy environments.

Similar Papers
  • Research Article
  • Citations52

Inferring Protein Sequence-Function Relationships with Large-Scale Positive-Unlabeled Learning.

  • Nov 18, 2020
  • Cell systems
  • Hyebin Song +4
  • Book Chapter
  • Citations6

A Tax Evasion Detection Method Based on Positive and Unlabeled Learning with Network Embedding Features

  • Jan 01, 2020
  • Lingyun Mi +3
  • Research Article
  • Citations81

Positive-unlabeled learning in bioinformatics and computational biology: a brief review.

  • Nov 03, 2021
  • Briefings in Bioinformatics
  • Fuyi Li +13
  • PDF
  • Research Article
  • Citations27

Embryo Grading With Unreliable Labels Due to Chromosome Abnormalities by Regularized PU Learning With Ranking.

  • Feb 01, 2022
  • IEEE Transactions on Medical Imaging
  • Masashi Nagaya +1
  • Conference Article
  • Citations10

Positive-Unlabeled Learning in the Face of Labeling Bias

  • Nov 01, 2015
  • Noah Youngs +2
  • Research Article
  • Citations61

AdaSampling for Positive-Unlabeled and Label Noise Learning With Bioinformatics Applications.

  • Apr 02, 2018
  • IEEE Transactions on Cybernetics
  • Pengyi Yang +5
  • Conference Article
  • Citations5

Risky Host Detection with Bias Reduced Semi-Supervised Learning

  • Jul 12, 2019
  • Shuning Wu +4
  • Conference Article
  • Citations49

Dist-PU: Positive-Unlabeled Learning from a Label Distribution Perspective

  • Jun 01, 2022
  • Yunrui Zhao +4
  • Conference Article
  • Citations2

Posistive-Unlabeled Learning via Optimal Transport and Margin Distribution

  • Jul 01, 2022
  • Nan Cao +3
  • Research Article
  • Citations9

False positive rate control for positive unlabeled learning

  • Aug 08, 2019
  • Neurocomputing
  • Shuchen Kong +5
  • Conference Article
  • Citations16

Learning from Multi-class Positive and Unlabeled Data

  • Nov 01, 2020
  • Senlin Shu +3
  • Conference Article
  • Citations10

Claim Verification under Positive Unlabeled Learning

  • Dec 07, 2020
  • Fan Yang +2
  • Conference Article
  • Citations8

Positive Unlabeled Learning with Class-prior Approximation

  • Jul 01, 2020
  • Shizhen Chang +2
  • Research Article
  • Citations3

Measuring the Significance of Policy Outputs with Positive Unlabeled Learning

  • Oct 19, 2020
  • American Political Science Review
  • Radoslaw Zubek +2
  • Research Article
  • Citations16

A novel landslide susceptibility prediction framework based on contrastive loss

  • Feb 02, 2024
  • GIScience & Remote Sensing
  • Shubing Ouyang +4
Cactus Communications logo

Copyright 2026 Cactus Communications. All rights reserved.