arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2103.16725cs.CVcs.LG
收藏

SimPLE:半监督分类的相似伪标签利用

SimPLE: Similar Pseudo Label Exploitation for Semi-Supervised Classification

University of Southern California(南加州大学)

Zijian Hu, Zhengyu Yang, Xuefeng Hu, Ram Nevatia

摘要:

一个常见的分类任务情况是,有大量的数据可用于训练,但只有一小部分用类标签注释。在这种情况下,半监督训练的目标是通过利用不仅来自标记数据而且来自大量未标记数据的信息来提高分类准确性。最近的工作已经开发了显着的改进,通过探索不同的增强标记和未标记的数据之间的一致性约束。沿着这条道路,我们提出了一种新的无监督目标,专注于研究彼此相似的高置信度未标记数据之间的关系。新提出的配对损失最大限度地减少了高置信度伪标签之间的统计距离,相似性高于一定的阈值。将配对损失与MixMatch家族开发的技术相结合,我们提出的SimPLE算法在CIFAR-100和Mini-ImageNet上显示出比以前的算法显着的性能增益,并且与CIFAR-10和SVHN上的最先进方法相当。此外,SimPLE在迁移学习设置中也优于最先进的方法,其中模型由ImageNet或DomainNet-Real上预先训练的权重初始化。该代码可在github.com/zijian-hu/SimPLE上获得。

英文摘要:

A common classification task situation is where one has a large amount of data available for training, but only a small portion is annotated with class labels. The goal of semi-supervised training, in this context, is to improve classification accuracy by leverage information not only from labeled data but also from a large amount of unlabeled data. Recent works have developed significant improvements by exploring the consistency constrain between differently augmented labeled and unlabeled data. Following this path, we propose a novel unsupervised objective that focuses on the less studied relationship between the high confidence unlabeled data that are similar to each other. The new proposed Pair Loss minimizes the statistical distance between high confidence pseudo labels with similarity above a certain threshold. Combining the Pair Loss with the techniques developed by the MixMatch family, our proposed SimPLE algorithm shows significant performance gains over previous algorithms on CIFAR-100 and Mini-ImageNet, and is on par with the state-of-the-art methods on CIFAR-10 and SVHN. Furthermore, SimPLE also outperforms the state-of-the-art methods in the transfer learning setting, where models are initialized by the weights pre-trained on ImageNet or DomainNet-Real. The code is available at github.com/zijian-hu/SimPLE.

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。

↑