L1 regularization is effective for feature selection, but the resulting optimization is challenging due to the non-differentiability of the 1-norm. In this paper we compare state...
This paper presents a decomposition method for efficiently constructing 1-norm Support Vector Machines (SVMs). The decomposition algorithm introduced in this paper possesses many d...
Many unsupervised algorithms for nonlinear dimensionality reduction, such as locally linear embedding (LLE) and Laplacian eigenmaps, are derived from the spectral decompositions o...
It is often the case that analyses of human interactive activity are lost once an article is written about the results obtained. Although it is clear that corpora are gathered in o...
We pose the problem of perfect segmentation for regions with ambiguous boundaries. We design machine learning classifiers to identify boundaries and build these into an interactiv...