Hyperbolic Hierarchical Clustering for Visual Representation Learning

Jianan Wei, Guikun Chen, Zhiyuan Weng, Yujia Wang, Wenguan Wang

Keywords: Clustering, representation learning, hyperbolic

Abstract:

We investigate the token mixer in vision backbones by revisiting clustering, one of the most classic approaches in machine learning.
An effective token mixer is a fundamental component of modern vision backbones like vision Transformers, facilitating information exchange between image patches. Mainstream token mixers, which rely on convolution, attention, MLP, or their hybrids, primarily focus on navigating the trade-off between accuracy and computational cost. However, a significant drawback of these methods is their black-box nature; their encoding process is opaque and lacks interpretability. Diverging from these opaque designs, we introduce ClusterMixer, a transparent token mixer that is grounded in a clustering paradigm and interpretable by design. ClusterMixer explicitly formulates the token mixing process through a hierarchical clustering mechanism. To model the natural, tree-like relationships inherent in visual data, the clustering is performed in hyperbolic space, which is well-suited for embedding hierarchies with low distortion. Building on this innovation, we present HCFormer, a new backbone architecture that integrates ClusterMixer with a series of meticulously designed clustering strategies to ensure robust performance across tasks. Extensive experiments demonstrate that HCFormer consistently outperforms its counterparts across diverse tasks, including image classification, object detection, instance segmentation, and semantic segmentation. Considering its transparency and efficacy, we hope HCFormer can facilitate a paradigm shift toward interpretable backbones. Our source code will be released.

Primary Area: unsupervised, self-supervised, semi-supervised, and supervised representation learning

Submission Number: 1256