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 2602.04255cs.LG
收藏

从模糊到行动:从POMDP视角看部分多标签模糊性及其一 horizon 解决方案

From Ambiguity to Action: A POMDP Perspective on Partial Multi-Label Ambiguity and Its Horizon-One Resolution

College of Computer Science and Technology, Jilin University, China(吉林大学计算机科学与技术学院);Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education, Jilin University, China(教育部符号计算与知识工程重点实验室)

Hanlin Pan, Yuhao Tang, Wanfu Gao

AI总结:

本文提出基于POMDP的框架,通过强化学习联合解决部分多标签学习中的标签消歧与特征选择问题,提升模型伪标签质量与特征可解释性。

摘要:

在部分多标签学习(PML)中,真实标签是不可观测的,这使得标签消歧变得重要但困难。一个关键挑战是模糊的候选标签可能会传播错误到下游任务,如特征工程。为了解决这个问题,我们将消歧和特征选择任务联合建模为部分可观测马尔可夫决策过程(POMDP),将PML风险最小化转化为期望回报最大化。第一阶段通过强化学习训练一个Transformer策略,生成高质量的硬伪标签;第二阶段将特征选择描述为一个序列强化学习问题,逐步选择特征并输出可解释的全局排名。我们进一步提供了PML-POMDP对应关系的理论分析以及将误差分解为伪标签质量和样本量的超额风险界。在多个指标和数据集上的实验验证了该框架的优势。

英文摘要:

In partial multi-label learning (PML), the true labels are unobserved, which makes label disambiguation important but difficult. A key challenge is that ambiguous candidate labels can propagate errors into downstream tasks such as feature engineering. To solve this issue, we jointly model the disambiguation and feature selection tasks as Partially Observable Markov Decision Processes (POMDP) to turn PML risk minimization into expected-return maximization. Stage 1 trains a transformer policy via reinforcement learning to produce high-quality hard pseudo-labels; Stage 2 describes feature selection as a sequential reinforcement learning problem, selecting features step by step and outputting an interpretable global ranking. We further provide the theoretical analysis of PML-POMDP correspondence and the excess-risk bound that decompose the error into pseudo label quality term and sample size. Experiments in multiple metrics and data sets verify the advantages of the framework.

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。

↑