Regularized Non-negative Spectral Embedding for Clustering
This addresses the problem of stage mismatch in spectral clustering for researchers and practitioners, offering an incremental improvement by integrating stages into a single framework.
The paper tackles the sub-optimal results in spectral clustering due to separate stages by proposing Regularized Non-negative Spectral Embedding (RNSE), an end-to-end method that adaptively learns similarity matrices and uses non-negative constraints for direct clustering, demonstrating superior performance on synthetic and real-world datasets.
Spectral Clustering is a popular technique to split data points into groups, especially for complex datasets. The algorithms in the Spectral Clustering family typically consist of multiple separate stages (such as similarity matrix construction, low-dimensional embedding, and K-Means clustering as post processing), which may lead to sub-optimal results because of the possible mismatch between different stages. In this paper, we propose an end-to-end single-stage learning method to clustering called Regularized Non-negative Spectral Embedding (RNSE) which extends Spectral Clustering with the adaptive learning of similarity matrix and meanwhile utilizes non-negative constraints to facilitate one-step clustering (directly from data points to clustering labels). Two well-founded methods, successive alternating projection and strategic multiplicative update, are employed to work out the quite challenging optimization problems in RNSE. Extensive experiments on both synthetic and real-world datasets demonstrate RNSE superior clustering performance to some state-of-the-art competitors.