arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2006.11488cs.LGstat.ML
收藏

从部分有效数据中恢复准确的标记信息以实现有效的多标记学习

Recovering Accurate Labeling Information from Partially Valid Data for Effective Multi-Label Learning

Jilin University(吉林大学);Hefei University of Technology(合肥工业大学)

Ximing Li, Yang Wang

摘要:

部分多标签学习(PML)旨在从带有噪声监督的数据集中诱导多标签预测器,其中每个训练实例与多个候选标签相关联,但仅部分有效。为了解决噪声问题,现有的PML方法基本上通过利用候选标签的地面实况置信度(\ie,候选标签是地面实况标签的可能性)来恢复地面实况标签。然而,它们忽略了来自非候选标签的信息,这可能有助于地面实况标签恢复。在本文中,我们提出了恢复地面真理标签,\ie估计地面真理的信心,从标签富集,组成的候选标签的相关度和非候选标签的无关度。在此基础上,我们进一步开发了一种新的两阶段PML方法,即\emph{\underline{P}artial \underline{M}ulti-\underline{L}abel \underline{L}earning with \underline{L}abel \underline{E}nrichment-\underline{R}ecovery}(\baby),其中在第一阶段,它估计标签富集与无约束的标签传播,然后联合学习地面真实的信心和多标签预测给定的标签富集。实验结果验证了\baby优于最先进的PML方法。

英文摘要:

Partial Multi-label Learning (PML) aims to induce the multi-label predictor from datasets with noisy supervision, where each training instance is associated with several candidate labels but only partially valid. To address the noisy issue, the existing PML methods basically recover the ground-truth labels by leveraging the ground-truth confidence of the candidate label, \ie the likelihood of a candidate label being a ground-truth one. However, they neglect the information from non-candidate labels, which potentially contributes to the ground-truth label recovery. In this paper, we propose to recover the ground-truth labels, \ie estimating the ground-truth confidences, from the label enrichment, composed of the relevance degrees of candidate labels and irrelevance degrees of non-candidate labels. Upon this observation, we further develop a novel two-stage PML method, namely \emph{\underline{P}artial \underline{M}ulti-\underline{L}abel \underline{L}earning with \underline{L}abel \underline{E}nrichment-\underline{R}ecovery} (\baby), where in the first stage, it estimates the label enrichment with unconstrained label propagation, then jointly learns the ground-truth confidence and multi-label predictor given the label enrichment. Experimental results validate that \baby outperforms the state-of-the-art PML methods.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑