Hierarchically Clustered Representation Learning
This work addresses the problem of learning hierarchical structures in embeddings for researchers in machine learning, though it appears incremental as it builds on prior joint optimization methods.
The paper tackled the limitation of flat clustering in representation learning by introducing Hierarchically-Clustered Representation Learning (HCRL), which jointly optimizes representation learning and hierarchical clustering, resulting in competent likelihoods and best accuracies in evaluations on image and text domains.
The joint optimization of representation learning and clustering in the embedding space has experienced a breakthrough in recent years. In spite of the advance, clustering with representation learning has been limited to flat-level categories, which often involves cohesive clustering with a focus on instance relations. To overcome the limitations of flat clustering, we introduce hierarchically-clustered representation learning (HCRL), which simultaneously optimizes representation learning and hierarchical clustering in the embedding space. Compared with a few prior works, HCRL firstly attempts to consider a generation of deep embeddings from every component of the hierarchy, not just leaf components. In addition to obtaining hierarchically clustered embeddings, we can reconstruct data by the various abstraction levels, infer the intrinsic hierarchical structure, and learn the level-proportion features. We conducted evaluations with image and text domains, and our quantitative analyses showed competent likelihoods and the best accuracies compared with the baselines.