Effective Tensor Sketching via Sparsification

10/31/2017
by   Dong Xia, et al.
0

In this paper, we investigate effective sketching schemes via sparsification for high dimensional multilinear arrays or tensors. More specifically, we propose a novel tensor sparsification algorithm that retains a subset of the entries of a tensor in a judicious way, and prove that it can attain a given level of approximation accuracy in terms of tensor spectral norm with a much smaller sample complexity when compared with existing approaches. In particular, we show that for a kth order d×...× d cubic tensor of stable rank r_s, the sample size requirement for achieving a relative error ε is, up to a logarithmic factor, of the order r_s^1/2 d^k/2 /ε when ε is relatively large, and r_s d /ε^2 and essentially optimal when ε is sufficiently small. It is especially noteworthy that the sample size requirement for achieving a high accuracy is of an order independent of k. To further demonstrate the utility of our techniques, we also study how higher order singular value decomposition (HOSVD) of large tensors can be efficiently approximated via sparsification.

READ FULL TEXT

Please sign up or login with your details

Forgot password? Click here to reset