arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2104.01429cs.CV
收藏

图对比聚类

Graph Contrastive Clustering

DAMO Academy, Alibaba Group(阿里巴巴达摩院);Peking University(北京大学);Shandong University(山东大学)

Huasong Zhong, Jianlong Wu, Chong Chen, Jianqiang Huang, Minghua Deng, Liqiang Nie, Zhouchen Lin, Xian-Sheng Hua

摘要:

最近,一些对比学习方法被提出来同时学习表示和聚类分配,取得了显着的改进。然而,这些方法没有考虑类别信息和聚类目标,因此学习的表示不是最佳的聚类和性能可能受到限制。针对这一问题,我们首先提出了一种新的图对比学习框架,然后将其应用于聚类任务,我们提出了图收缩聚类~(GCC)方法。与基本的对比聚类不同,它只假设图像及其增强应该共享相似的表示和聚类分配,我们提升了实例级的一致性,假设一个聚类中的样本及其增强都应该是相似的。具体来说,一方面,提出了基于图拉普拉斯算子的对比损失来学习更具区分性和聚类友好的特征。另一方面,提出了一种新的基于图的对比学习策略来学习更紧凑的聚类分配。这两种方法都引入了潜在的类别信息,以减少类内方差,同时增加类间方差。六个常用的数据集上的实验证明了我们所提出的方法比最先进的方法的优越性。

英文摘要:

Recently, some contrastive learning methods have been proposed to simultaneously learn representations and clustering assignments, achieving significant improvements. However, these methods do not take the category information and clustering objective into consideration, thus the learned representations are not optimal for clustering and the performance might be limited. Towards this issue, we first propose a novel graph contrastive learning framework, which is then applied to the clustering task and we come up with the Graph Constrastive Clustering~(GCC) method. Different from basic contrastive clustering that only assumes an image and its augmentation should share similar representation and clustering assignments, we lift the instance-level consistency to the cluster-level consistency with the assumption that samples in one cluster and their augmentations should all be similar. Specifically, on the one hand, the graph Laplacian based contrastive loss is proposed to learn more discriminative and clustering-friendly features. On the other hand, a novel graph-based contrastive learning strategy is proposed to learn more compact clustering assignments. Both of them incorporate the latent category information to reduce the intra-cluster variance while increasing the inter-cluster variance. Experiments on six commonly used datasets demonstrate the superiority of our proposed approach over the state-of-the-art methods.

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。

↑