Unsupervised spectral learning
This addresses the need for unsupervised similarity learning in spectral clustering, which is incremental as it builds on existing methods by automating similarity construction.
The paper tackles the problem of constructing similarity matrices for spectral clustering without supervision, proposing an algorithm that learns a parametric similarity function from pairwise features while simultaneously clustering the data, and reports promising results on synthetic and real data.
In spectral clustering and spectral image segmentation, the data is partioned starting from a given matrix of pairwise similarities S. the matrix S is constructed by hand, or learned on a separate training set. In this paper we show how to achieve spectral clustering in unsupervised mode. Our algorithm starts with a set of observed pairwise features, which are possible components of an unknown, parametric similarity function. This function is learned iteratively, at the same time as the clustering of the data. The algorithm shows promosing results on synthetic and real data.