通过学习类别间关系提升未知类别的发现能力
ReLKD: Inter-Class Relation Learning with Knowledge Distillation for Generalized Category Discovery
- 设计双粒度模块，分别捕捉类别细粒度特征与粗粒度层级关系
- 利用知识蒸馏将粗粒度关系知识注入细粒度分类器，提升新类别识别
- 在小样本标注场景下表现优异，适合标注数据稀缺的发现任务
广义类别发现（GCD）需在仅提供已知类别标签的情况下，对包含已知和未知类别的无标签数据进行分类。以往方法多独立处理各类别，忽视了类别间的内在关联。实际场景中直接获取此类关系极为困难。为此，本文提出ReLKD，一个端到端框架，有效挖掘隐含的类别间关系，并利用该知识增强对新类别的分类性能。ReLKD包含三个核心模块：目标粒度模块用于学习判别性表征，粗粒度模块用于捕获层级化类别关系，蒸馏模块则将粗粒度模块的知识迁移至目标粒度模块，以优化其表征学习。在四个数据集上的大量实验表明，ReLKD在标注数据有限的场景下尤为有效。代码已公开于https://github.com/ZhouF-ECNU/ReLKD。
原文摘要 · Abstract (English)
Generalized Category Discovery (GCD) faces the challenge of categorizing unlabeled data containing both known and novel classes, given only labels for known classes. Previous studies often treat each class independently, neglecting the inherent inter-class relations. Obtaining such inter-class relations directly presents a significant challenge in real-world scenarios. To address this issue, we propose ReLKD, an end-to-end framework that effectively exploits implicit inter-class relations and leverages this knowledge to enhance the classification of novel classes. ReLKD comprises three key modules: a target-grained module for learning discriminative representations, a coarse-grained module for capturing hierarchical class relations, and a distillation module for transferring knowledge from the coarse-grained module to refine the target-grained module's representation learning. Extensive experiments on four datasets demonstrate the effectiveness of ReLKD, particularly in scenarios with limited labeled data. The code for ReLKD is available at https://github.com/ZhouF-ECNU/ReLKD.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。