tensor-decomposition v0.0.1

Higher order tensor decomposition

Implements the symmetric higher order power method (S-HOPM) and uses it to find non-negative rank-one decompositions of symmetric higher order (real) tensors. This is a very specific case of one possible generalization of the eigendecomposition of a matrix.

Note that apart from being limited to non-negative decompositions of symmetric tensors, the current version is also limited to tensors based on a 2D vector space (there is no limit on the order/degree of the tensors).

The use of the S-HOPM and its convergence properties is based on:

On the Best Rank-1 Approximation of Higher-Order Supersymmetric Tensors SIAM Journal on Matrix Analysis and Applications, Vol. 23, No. 3. (January 2002), pp. 863-884, doi:10.1137/s0895479801387413 by Eleftherios Kofidis and Phillip A. Regalia