发表机构
College of Systems Engineering, National University of Defense Technology（国防科技大学系统工程学院）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
针对标签噪声下的前向校正分类，提出选择性后验边际正则化（SPMR），利用反向后验冲突进行分级更新，在多个基准上显著提升分类性能。
AI 中文摘要
基于类条件标签噪声的学习通常依赖于从潜在干净类别到观测注释的转移模型。前向校正将这种转移嵌入似然函数中，然而有限样本的网络仍可能记忆损坏的标签。校正后的似然还引入了一个关于干净类别的反向后验，该后验可以解释每个注释。当其主导类别与注释不同时，模型和转移矩阵提供了反对该注释的证据，但主导的替代类别可能几乎持平。我们引入了选择性后验边际正则化（SPMR），它保留了前向目标，并将这种不一致转化为对干净分类器的分级更新。SPMR选择主导的反向后验类别，通过两个主导后验类别之间的分离度来缩放一个分离的成对边际，并对分散的冲突赋予相应较小的影响。该间隙分解为转移调整的成对分离和主导对携带的后验质量。主动边际遵循局部最小范数logit方向，该方向扩大所选的成对边际。在五个已知转移基准上，SPMR将完整长度的前向方法提高了2.5-7.0个百分点，并且比带有Mixup和早停的前向方法高出0.7-2.5个百分点。匹配的干预措施支持后验空间系数、转移调整目标和成对行动的独特收益。相同的设计可转移到估计的转移、人工注释、架构更改和更强的前向配方。该公式利用了前向校正中已有的潜在类别证据，而不将每个后验冲突提升为校正标签。
英文摘要
Learning with class-conditional label noise often relies on a transition model from latent clean classes to observed annotations. Forward correction embeds this transition in the likelihood, yet finite-sample networks may still memorize corrupted labels. The corrected likelihood also induces a reverse posterior over the clean classes that could explain each annotation. When its leading class differs from the annotation, the model and transition matrix provide evidence against that annotation, but the leading alternatives can remain nearly tied. We introduce Selective Posterior Margin Regularization (SPMR), which preserves the Forward objective and converts this disagreement into a graded update on the clean classifier. SPMR selects the leading reverse-posterior class, scales a detached pairwise margin by the separation between the two leading posterior classes, and assigns correspondingly little influence to diffuse conflicts. The gap factorizes into transition- adjusted pairwise separation and the posterior mass carried by the leading pair. The active margin follows the locally minimum-norm logit direction that enlarges the selected pairwise margin. Across five known-transition benchmarks, SPMR improves full-length Forward by 2.5-7.0 percentage points and remains 0.7-2.5 percentage points above Forward with Mixup and early stopping. Matched interventions support distinct gains from the posterior-space coefficient, transition-adjusted target, and pairwise action. The same design transfers to estimated transitions, human annotations, architectural changes, and stronger Forward recipes. The formulation uses latent-class evidence already available inside Forward correction without promoting every posterior conflict to a corrected label.
