Deep Clustering with Self-Supervision using Pairwise Similarities
This work addresses clustering challenges in machine learning by enhancing accuracy, though it appears incremental as it builds on existing deep clustering methods.
The paper tackles the problem of deep clustering by proposing a two-phase framework that first forms hypersphere-like groups in latent space and then uses pairwise similarities to handle complex distributions, achieving improved clustering performance on seven benchmark datasets.
Deep clustering incorporates embedding into clustering to find a lower-dimensional space appropriate for clustering. In this paper, we propose a novel deep clustering framework with self-supervision using pairwise similarities (DCSS). The proposed method consists of two successive phases. In the first phase, we propose to form hypersphere-like groups of similar data points, i.e. one hypersphere per cluster, employing an autoencoder that is trained using cluster-specific losses. The hyper-spheres are formed in the autoencoder's latent space. In the second phase, we propose to employ pairwise similarities to create a $K$-dimensional space that is capable of accommodating more complex cluster distributions, hence providing more accurate clustering performance. $K$ is the number of clusters. The autoencoder's latent space obtained in the first phase is used as the input of the second phase. The effectiveness of both phases is demonstrated on seven benchmark datasets by conducting a rigorous set of experiments.