arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2209.14385cs.CVcs.AIcs.CRcs.LG
收藏

开集识别自监督表示学习中的特征解耦

Feature Decoupling in Self-supervised Representation Learning for Open Set Recognition

Florida Institute of Technology(佛罗里达理工学院)

Jingyun Jia, Philip K. Chan

摘要:

假设在分类过程中可能存在未知类,开集识别(OSR)任务的目标是将实例分类到已知类或将其拒绝为未知类。在本文中,我们使用两个阶段的训练策略的OSR问题。在第一阶段中,我们引入了一种自监督特征解耦方法,该方法从已知类中找到输入样本的内容特征。具体来说,我们的特征解耦方法学习了一种可以分为内容特征和转换特征的表示。在第二阶段,我们用类标签微调内容特征。然后将微调的内容特性用于OSR问题。此外,我们考虑了一个无监督的OSR场景,在那里我们聚类从第一阶段学习的内容特征。为了衡量表示质量,我们引入了帧内-帧间比(IIR)。我们的实验结果表明,我们提出的自监督方法优于其他图像和恶意软件OSR问题。此外,我们的分析表明,IIR与OSR性能相关。

英文摘要:

Assuming unknown classes could be present during classification, the open set recognition (OSR) task aims to classify an instance into a known class or reject it as unknown. In this paper, we use a two-stage training strategy for the OSR problems. In the first stage, we introduce a self-supervised feature decoupling method that finds the content features of the input samples from the known classes. Specifically, our feature decoupling approach learns a representation that can be split into content features and transformation features. In the second stage, we fine-tune the content features with the class labels. The fine-tuned content features are then used for the OSR problems. Moreover, we consider an unsupervised OSR scenario, where we cluster the content features learned from the first stage. To measure representation quality, we introduce intra-inter ratio (IIR). Our experimental results indicate that our proposed self-supervised approach outperforms others in image and malware OSR problems. Also, our analyses indicate that IIR is correlated with OSR performance.

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。

↑