Tensor Regression Techniques

We include three major functions:

1. tensorreg: Tensor regression method with known basis for decomposition
2. tucker_onesetp: Tucker Decomposition Regression, Use Tucker decomposition to learn basis and then perform regression.

Other utility function include:

• bsplineBasis: Create B-spline basis

• matricize: matricize tensor along certain dimension.

• createSigma: a method to create covariance matrix, used to create non-i.i.d case.