Spectral Clustering with Jensen-type kernels and their multi-point extensions
This work addresses the need for more flexible similarity measures in clustering for machine learning applications, though it appears incremental as it extends existing kernel methods to multi-point cases.
The authors tackled the problem of measuring similarity among multiple points in spectral clustering by proposing multi-point kernels based on Jensen-type divergences, developing a multi-point spectral clustering method with tensor flattening, and demonstrating its effectiveness on standard datasets and image segmentation tasks.
Motivated by multi-distribution divergences, which originate in information theory, we propose a notion of `multi-point' kernels, and study their applications. We study a class of kernels based on Jensen type divergences and show that these can be extended to measure similarity among multiple points. We study tensor flattening methods and develop a multi-point (kernel) spectral clustering (MSC) method. We further emphasize on a special case of the proposed kernels, which is a multi-point extension of the linear (dot-product) kernel and show the existence of cubic time tensor flattening algorithm in this case. Finally, we illustrate the usefulness of our contributions using standard data sets and image segmentation tasks.