arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2112.03398cs.LGcs.CV
收藏

基于多生成器GANs的自顶向下深度聚类

Top-Down Deep Clustering with Multi-generator GANs

Department of Computer Science, Universidade Federal de Minas Gerais(米纳斯吉拉斯联邦大学计算机科学系);ESRI Inc.(环境系统研究所公司)

Daniel de Mello, Renato Assunção, Fabricio Murai

摘要:

深度聚类(DC)利用深度架构的表示能力来学习最适合聚类分析的嵌入空间。这种方法过滤掉了与聚类无关的低级信息,并已被证明在高维数据空间中非常成功。一些DC方法采用生成对抗网络(GANs),其动机是这些模型能够隐式学习的强大潜在表示。在这项工作中,我们提出了HC-MGAN,这是一种基于具有多个生成器(MGANs)的GANs的新技术,尚未探索用于聚类。我们的方法的灵感来自于观察,每个生成器的MGAN往往生成的数据,与子区域的真实数据分布。我们使用这个集群生成训练分类器,用于推断给定图像来自哪个生成器,从而为真实分布提供语义上有意义的聚类。此外,据我们所知,我们设计了我们的方法,以便在自上而下的分层聚类树中执行,从而提出了第一个分层DC方法。我们进行了几个实验,以评估所提出的方法对最近的DC方法,获得竞争力的结果。最后,我们进行了探索性分析的层次聚类树,突出如何准确地组织数据的层次结构的语义连贯的模式。

英文摘要:

Deep clustering (DC) leverages the representation power of deep architectures to learn embedding spaces that are optimal for cluster analysis. This approach filters out low-level information irrelevant for clustering and has proven remarkably successful for high dimensional data spaces. Some DC methods employ Generative Adversarial Networks (GANs), motivated by the powerful latent representations these models are able to learn implicitly. In this work, we propose HC-MGAN, a new technique based on GANs with multiple generators (MGANs), which have not been explored for clustering. Our method is inspired by the observation that each generator of a MGAN tends to generate data that correlates with a sub-region of the real data distribution. We use this clustered generation to train a classifier for inferring from which generator a given image came from, thus providing a semantically meaningful clustering for the real distribution. Additionally, we design our method so that it is performed in a top-down hierarchical clustering tree, thus proposing the first hierarchical DC method, to the best of our knowledge. We conduct several experiments to evaluate the proposed method against recent DC methods, obtaining competitive results. Last, we perform an exploratory analysis of the hierarchical clustering tree that highlights how accurately it organizes the data in a hierarchy of semantically coherent patterns.

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。

↑