We investigate the problem of learning a classification task for datasets which are described by matrices. Rows and columns of these matrices correspond to objects, where row and column objects may belong to different sets, and the entries in the matrix express the relationships between them. We interpret the matrix elements as being produced by an unknown kernel which operates on object pairs ...

structures on a model based on available prior informaAlgorithms are presented for least-squares approximation. Earlier work on these problems has primarily contion of Toeplitz and Hankel matrices from noise corsisted of using Singular Value Decomposition [4, 5, 81, rupted or ill-composed matrices, which may not have where only the rank information of the underlying sigcorrect structural or ran...