Abstract Deep clustering has been dominated by flat models that split a dataset into a predefined number of groups. Although recent methods achieve an extremely high similarity to the ground truth on popular benchmarks, the information contained in a flat partition is limited. In this paper, we introduce CoHiClust, a Contrastive Hierarchical Clustering model based on deep neural networks, which can be applied to typical image data. By employing a self-supervised learning approach, CoHiClust distills the base network into a binary tree without access to any labeled data. The hierarchical clustering structure can be used to analyze the relationship between clusters, as well as to measure the similarity between data points. In addition to the hierarchical structure we propose two visualization techniques, which allow us to deliver an intuitive explanation of tree nodes. Experiments demonstrate that CoHiClust generates a reasonable structure of clusters, which is consistent with our intuition and image semantics. Moreover, it obtains superior clustering accuracy on most of the image datasets compared to the state-of-the-art flat clustering models.
Pach et al. (Thu,) studied this question.