Learning With Proper Partial Labels

Author:

Wu Zhenguo1,Lv Jiaqi2,Sugiyama Masashi34

Affiliation:

1. University of Tokyo, Bunkyo, Tokyo 113-0033, Japan zhenguo@ms.k.u-tokyo.ac.jp

2. RIKEN AIP, Tokyo 103-0027, Japan jiaqi.lyu@riken.jp

3. RIKEN AIP, Tokyo 103-0027, Japan

4. University of Tokyo, Bunkyo, Tokyo 113-0033, Japan sugi@k.u-tokyo.ac.jp

Abstract

Abstract Partial-label learning is a kind of weakly supervised learning with inexact labels, where for each training example, we are given a set of candidate labels instead of only one true label. Recently, various approaches on partial-label learning have been proposed under different generation models of candidate label sets. However, these methods require relatively strong distributional assumptions on the generation models. When the assumptions do not hold, the performance of the methods is not guaranteed theoretically. In this letter, we propose the notion of properness on partial labels. We show that this proper partial-label learning framework requires a weaker distributional assumption and includes many previous partial-label learning settings as special cases. We then derive a unified unbiased estimator of the classification risk. We prove that our estimator is risk consistent, and we also establish an estimation error bound. Finally, we validate the effectiveness of our algorithm through experiments.

Publisher

MIT Press

Subject

Cognitive Neuroscience,Arts and Humanities (miscellaneous)

Reference35 articles.

1. Classification from pairwise similarity and unlabeled data;Bao,2018

2. Confidence scores make instance-dependent label-noise learning possible;Berthon,2021

3. Learning from similarity-confidence data;Cao,2021

4. On symmetric losses for learning from corrupted labels;Charoenphakdee,2019

Cited by 3 articles. 订阅此论文施引文献 订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献

1. Towards Effective Visual Representations for Partial-Label Learning;2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR);2023-06

2. Partial label learning: Taxonomy, analysis and outlook;Neural Networks;2023-04

3. On the Robustness of Average Losses for Partial-Label Learning;IEEE Transactions on Pattern Analysis and Machine Intelligence;2023

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.5亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2024 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3