Clustering-friendly Representation Learning for Enhancing Salient Features
This work addresses the challenge of tailoring unsupervised representation learning for specific downstream tasks like image clustering, though it appears incremental as it extends existing contrastive learning and analysis approaches.
The paper tackles the problem of unsupervised representation learning methods being unable to distinguish important features for specific downstream tasks like image clustering, and proposes a method that enhances clustering-critical features using contrastive analysis with a reference dataset, achieving higher clustering scores across three datasets compared to conventional methods.
Recently, representation learning with contrastive learning algorithms has been successfully applied to challenging unlabeled datasets. However, these methods are unable to distinguish important features from unimportant ones under simply unsupervised settings, and definitions of importance vary according to the type of downstream task or analysis goal, such as the identification of objects or backgrounds. In this paper, we focus on unsupervised image clustering as the downstream task and propose a representation learning method that enhances features critical to the clustering task. We extend a clustering-friendly contrastive learning method and incorporate a contrastive analysis approach, which utilizes a reference dataset to separate important features from unimportant ones, into the design of loss functions. Conducting an experimental evaluation of image clustering for three datasets with characteristic backgrounds, we show that for all datasets, our method achieves higher clustering scores compared with conventional contrastive analysis and deep clustering methods.