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 2605.07557cs.LG
收藏

超越分布估计:基于简单集锚定的结构推断以实现通用半监督学习

Beyond Distribution Estimation: Simplex Anchored Structural Inference Towards Universal Semi-Supervised Learning

Key Laboratory of New Generation Artificial Intelligence Technology and Its Interdisciplinary Applications (Southeast University), Ministry of Education, Nanjing, China(新一代人工智能技术及其交叉应用重点实验室(东南大学),教育部,南京,中国);School of Computer Science and Engineering, Southeast University, Nanjing, China(计算机科学与工程学院,东南大学,南京,中国);School of Electrical Engineering, Southeast University, Nanjing, China(电气工程学院,东南大学,南京,中国)

Yaxin Hou, Jun Ma, Hanyang Li, Bo Han, Jie Yu, Yuheng Jia

AI总结:

本文提出SAGE方法,通过在表示层面进行结构推断,避免分布估计,利用简单集锚定框架提升半监督学习性能,实验表明其在五个基准数据集上平均准确率提升8.52%。

摘要:

半监督学习在现实场景中面临挑战,因为标记数据稀缺且未标记数据遵循未知任意分布。我们正式将这一关键但研究不足的范式定义为通用半监督学习(UniSSL)。现有方法通常通过伪标签利用未标记数据,但它们通常依赖于理想化的均匀未标记数据分布假设或需要足够的标记数据来估计该分布。在UniSSL设置中,这种依赖性导致大量错误的伪标签,从而引发表示混淆。幸运的是,我们观察到由表示捕获的样本间关系比伪标签更可靠。利用这一见解,我们专注于表示层面的结构推断以绕过分布估计。相应地,我们提出了简单集锚定图-状态等分(SAGE),该方法捕捉高阶样本间依赖关系,以建立结构共识以指导表示学习。同时,为了缓解表示混淆,我们采用满足简单集等角紧框架的向量作为坐标框架,以指导类间表示分离。最后,我们引入基于分布无关度量的加权策略,以优先考虑可靠的伪标签,并引入一个辅助分支以隔离可能的错误伪标签。在五个标准基准上的评估表明,SAGE在五个标准基准上始终优于最先进的方法,平均准确率提升8.52%。

英文摘要:

Semi-supervised learning faces significant challenges in realistic scenarios where labeled data is scarce and unlabeled data follows unknown, arbitrary distributions. We formalize this critical yet under-explored paradigm as Universal Semi-supervised Learning (UniSSL). Existing methods typically leverage unlabeled data via pseudo-labeling. However, they often rely on the idealized assumption of a uniform unlabeled data distribution or require sufficient labeled data to estimate it. In the UniSSL setting, such dependencies lead to numerous erroneous pseudo-labels, thereby triggering representation confusion. Fortunately, we observe that inter-sample relations captured by representations are more reliable than pseudo-labels. Leveraging this insight, we shift our focus to representation-level structural inference to bypass distribution estimation. Accordingly, we propose Simplex Anchored Graph-state Equipartition (SAGE), which captures high-order inter-sample dependencies to establish structural consensus for guiding representation learning. Meanwhile, to mitigate representation confusion, we employ vectors that satisfy a simplex equiangular tight frame to serve as a coordinate frame for guiding inter-class representation separation. Finally, we introduce a weighting strategy based on distribution-agnostic metrics to prioritize reliable pseudo-labels and an auxiliary branch to isolate potentially erroneous pseudo-labels. Evaluations on five standard benchmarks show that SAGE consistently outperforms state-of-the-art methods, with an average accuracy gain of $\textbf{8.52%}$.

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。

↑