Subquadratic Kronecker Regression with Applications to Tensor Decomposition
share
Kronecker regression is a highly-structured least squares problem min_𝐱‖𝐊𝐱 - 𝐛‖_2^2, where the design matrix 𝐊 = 𝐀^(1)⊗⋯⊗𝐀^(N) is a Kronecker product of factor matrices. This regression problem arises in each step of the widely-used alternating least squares (ALS) algorithm for computing the Tucker decomposition of a tensor. We present the first subquadratic-time algorithm for solving Kronecker regression to a (1+ε)-approximation that avoids the exponential term O(ε^-N) in the running time. Our techniques combine leverage score sampling and iterative methods. By extending our approach to block-design matrices where one block is a Kronecker product, we also achieve subquadratic-time algorithms for (1) Kronecker ridge regression and (2) updating the factor matrix of a Tucker decomposition in ALS, which is not a pure Kronecker regression problem, thereby improving the running time of all steps of Tucker ALS. We demonstrate the speed and accuracy of this Kronecker regression algorithm on synthetic data and real-world image tensors.
READ FULL TEXT
