Representation Learning for Clustering via Building Consensus

TL;DR

This paper introduces ConCURL, a novel unsupervised learning framework that incorporates consensus consistency to improve image clustering performance across multiple datasets and conditions.

Contribution

It proposes a new consensus consistency notion and integrates it with exemplar and population consistencies in an end-to-end framework for enhanced clustering.

Findings

Outperforms state-of-the-art methods on four of five datasets

Extends evaluation to real-world distribution shifts

Provides comprehensive ablation studies

Abstract

In this paper, we focus on unsupervised representation learning for clustering of images. Recent advances in deep clustering and unsupervised representation learning are based on the idea that different views of an input image (generated through data augmentation techniques) must be close in the representation space (exemplar consistency), and/or similar images must have similar cluster assignments (population consistency). We define an additional notion of consistency, consensus consistency, which ensures that representations are learned to induce similar partitions for variations in the representation space, different clustering algorithms or different initializations of a single clustering algorithm. We define a clustering loss by executing variations in the representation space and seamlessly integrate all three consistencies (consensus, exemplar and population) into an end-to-end learning framework. The proposed algorithm, consensus clustering using unsupervised representation learning (ConCURL), improves upon the clustering performance of state-of-the-art methods on four out of five image datasets. Furthermore, we extend the evaluation procedure for clustering to reflect the challenges encountered in real-world clustering tasks, such as maintaining clustering performance in cases with distribution shifts. We also perform a detailed ablation study for a deeper understanding of the proposed algorithm. The code and the trained models are available at https://github.com/JayanthRR/ConCURL_NCE.